Materials and Methods

Animals and training—This study was approved by the Committee for Animal Welfare at the Faculty of Veterinary Medicine, Utrecht University. Ten Standardbred geldings with a mean ± SD age of 20 ± 2 months and weight of 384 ± 42 kg (range, 331 to 485 kg) were used in this study. Horses were individually housed. Their diet consisted of grass silage supplemented with concentrate feed according to the daily estimated energy requirements of 58 MJ of net energy (range, 54 to 66 MJ of net energy; Appendix 1). Salt blocks and water were available ad libitum. During the intensified training period, a supplement with vitamin E, selenium, electrolytes, and other vitamins was added to the diet of all horses.^a,b All horses had a body condition score between 4 and 5 (out of a possible score of 9) during the study.¹⁵

Prior to commencing training, horses were acclimated to exercising on a treadmill^c for 4 weeks. They were trained for 24 weeks and detrained for 4 weeks. All training and exercise tests took place on a high-speed treadmill. At the end of each training phase, an SET was performed at approximately 80% of the HR_max (equivalent to a treadmill speed to 7.5 to 8.5 m/s with a 1% to 4% incline) for 20 minutes. Prior to SETs, food was withheld from horses for approximately 1 hour. Horses were paired on the basis of age. From each pair, 1 horse was randomly assigned to the intensified training program and the other horse served as an agematched control and continued the established training program.

Monitoring training—To standardize training to the exercise capacity of each horse, training and exercise intensity were adapted to the HR_max of each horse on the basis of successful completion of an incremental exercise test on the high-speed treadmill, in which a plateau in HR_max was achieved. The incremental exercise test started with a warm-up, and horses then trotted for 2 minutes at 5 m/s, followed by 2 minutes at 6 m/s. Intensity was increased by 1 m/s every 2 minutes until the horses reached fatigue, which was defined as the speed at which horses could not keep up with the treadmill despite humane encouragement. Heart rate was monitored with an online ECG recording.^d,e

Training was divided into 4 phases. During phase 1, horses were introduced to the high-speed treadmill for 4 weeks. Each exercise session was preceded by a 30-minute warm-up by use of a mechanical rotating (hot) walker, followed by an 8-minute warm-up on the high-speed treadmill, which consisted of 4 minutes at 1.6 m/s and 4 minutes at 3.0 to 4.0 m/s, with no incline. The program during phase 1 consisted of the following endurance exercises: week 1, 30% of HR_max for 20 to 30 minutes 3 d/wk; week 2, 30% of HR_max for 25 to 45 minutes 4 d/wk; week 3, 40% of HR_max for 30 to 45 minutes 4 d/wk; and week 4, 50% of HR_max for 35 to 45 minutes 4 d/wk. Each exercise session ended with a cool-down period, consisting of a 5-minute walk on the treadmill followed by a 30-minute walk on the hot walker.

During phase 2, horses received an 18-week training program of mixed endurance training and high-intensity exercise training.¹⁶ Days of endurance training were alternated with days of high-intensity exercise training. Each training session was preceded by a 30-minute warm-up on the hot walker followed by an 8-minute warm-up (4 minutes at 1.6 m/s followed by 4 minutes at 4.5 m/s) on the treadmill. Endurance training included 20 to 24 minutes of continuous running at 60% of HR_max or 16 to 18 minutes at 75% of HR_max. The interval training included three 3-minute bouts at 80% to 85% of HR_max or four 2-minute bouts at 80% to 85% of HR_max interspersed with 3- or 2-minute periods at 60% of HR_max. Each training session ended with a cool-down period consisting of a 5-minute walk on the treadmill, followed by 30 minutes of walking on the hot walker. Horses exercised 4 d/wk throughout phase 2.

During phase 3, control horses continued training at the same level as in phase 2 for 6 weeks, whereas IT horses underwent alternating days of high-intensity exercise training and endurance training for 6 d/wk during the first 3 weeks of phase 3, followed by 3 weeks of training for 7 d/wk according to a previously validated protocol in Standardbreds for overtraining characterized by abolishment of resting days.¹⁶ Each training session was preceded by a 30-minute warm-up period on the hot walker followed by an 8-minute warm-up (4 minutes at 1.6 m/s followed by 4 minutes at 4.5 m/s) on the treadmill. Exercise intensity and duration for endurance training were gradually increased to 24 to 35 minutes at 60% to 75% of HR_max. High-intensity exercise gradually increased to five 3-minute bouts at 80% to 85% of HR_max interspersed with 2-minute periods at 60% of HR_max or six 2-minute bouts at 80% to 85% of HR_max interspersed with 1-minute or 2-minute periods at 60% of HR_max.

In phase 4, horses received a 4-week deconditioning program of light endurance exercise. Horses performed endurance training for 20 minutes at 60% of HR_max for 3 days and 70% of HR_max for 1 day a week.

On the resting days in all phases, horses walked for 60 minutes on the hot walker. At the end of each phase, an SET was performed. Heart rate during the SET was monitored as during the incremental exercise test. Horses were trained on the basis of individual fitness, the goal being to keep horses at a work rate of approximately 80% of their HR_max. Heart rate was monitored daily to ensure that horses were working at the proper level, and every week their program was adjusted (speed or duration) on the basis of those results.

Blood sample collection—Blood samples were collected via jugular venipuncture immediately before and 60 minutes after each SET. For the analysis of BHBA, nonesterified fatty acids, total free fatty acids, and lactic acid, blood was collected into tubes containing lithium heparin (for analysis of free fatty acids, nonesterified fatty acids, and lactic acid) and serum clot tubes (for analysis of BHBA) before, immediately after, and 60 minutes after ending the SET. Blood for analysis of lactic acid was also obtained during the SET at 9, 14, and 19 minutes. Blood samples were centrifuged^f for 10 minutes at 6,000 × g; plasma was harvested, stored at 20°C, and analyzed after termination of the study. Blood for analysis of lactic acid was stored on ice and analyzed immediately.

Biochemical analysis—Identification analyses were performed by use of gas chromatography–mass spectrometry on a gas chromatograph^g linked to a mass spectrometer.^h Prior to gas chromatography–mass spectrometry analysis, fatty acids were trimethylsilylated with N,N-bis (trimethylsilyl) trifluoracetamide-pyridine-trimethylchlorosilane mixture (5:1:0.05 vol/vol/ vol) at 60°C for 30 minutes. The gas chromatographic separation was performed on a 25-m × 0.25-mm capillary column (film thickness, 0.19 mm).ⁱ

Free carnitine and acylcarnitines in plasma were analyzed as their butyl ester derivatives by use of electrospray tandem mass spectrometry on a micromass system^j equipped with a high-performance liquid chromatography system.^k The coefficient of variation for the determination of acylcarnitine concentrations was 10% to 15%.

Plasma total free fatty acid concentrations were measured by use of a commercial kit^l that had been validated for use in samples obtained from horses. The intra-assay and interassay coefficients of variation were 7.3% and 16.0%, respectively. Curves obtained with serial dilutions were parallel to the standard curve.

Serum BHBA concentrations were measured by use of a commercial kit^m that had been validated for use in samples obtained from horses. The intra-assay and interassay coefficients of variation were 7.8% and 10.9%, respectively. Curves obtained with serial dilutions were parallel to the standard curve. Plasma lactic acid concentration was determined by use of an automatic analyzer.ⁿ

Statistical analysis—A commercially available software program was used for analyses.^o Results during training and deconditioning were compared by use of a 2-way repeated-measures ANOVA, with training group as between subject variable and before versus after and time as a repeated-measures factor. Post hoc comparisons were made by use of a Bonferroni test when F values were significant (P < 0.05). Significant differences in trotting time during the phase 3 SET between groups were assessed by use of the independent t test. Results are presented as mean ± SD values. Values of P < 0.05 were considered significant.

Results

The first exercise test at the end of the acclimatization period was difficult to perform for the young minimally trained horses. Mean plasma lactic acid concentration during this first exercise test was less than the threshold of 4 mmol/L, and as a consequence, the adaptation period was not regarded as training per se.

A right shift of the lactic acid concentration versus time curve during the phase 3 SET, compared with that of the phase 2 SET, was evident in both horse groups (Figure 1). The right shift of the lactic acid concentration versus time curve was still evident during the phase 4 SET for IT horses, but not for control horses.

Mean plasma lactic acid concentration versus time curves of the phase 2 (SET 2), phase 3 (SET 3), and phase 4 (SET 4) SETs during the first 10 minutes of trot in control horses (A) or IT horses (B). The first 9 minutes of the SET consisted of a warm-up period. A right shift of the lactic acid concentration versus time curve during the phase 3 SET, compared with phase 2 SET, is seen for both horse groups. The right shift of the lactic acid concentration versus time curve can still be seen during the phase 4 SET for IT horses and not for control horses.

Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1469

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Mean plasma lactic acid concentration versus time curves of the phase 2 (SET 2), phase 3 (SET 3), and phase 4 (SET 4) SETs during the first 10 minutes of trot in control horses (A) or IT horses (B). The first 9 minutes of the SET consisted of a warm-up period. A right shift of the lactic acid concentration versus time curve during the phase 3 SET, compared with phase 2 SET, is seen for both horse groups. The right shift of the lactic acid concentration versus time curve can still be seen during the phase 4 SET for IT horses and not for control horses.

Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1469

• Download Figure
• Download figure as PowerPoint slide

Mean plasma lactic acid concentration versus time curves of the phase 2 (SET 2), phase 3 (SET 3), and phase 4 (SET 4) SETs during the first 10 minutes of trot in control horses (A) or IT horses (B). The first 9 minutes of the SET consisted of a warm-up period. A right shift of the lactic acid concentration versus time curve during the phase 3 SET, compared with phase 2 SET, is seen for both horse groups. The right shift of the lactic acid concentration versus time curve can still be seen during the phase 4 SET for IT horses and not for control horses.

Citation: American Journal of Veterinary Research 69, 11; 10.2460/ajvr.69.11.1469

• Download Figure
• Download figure as PowerPoint slide

Plasma free fatty acid and acylcarnitine ester concentrations were determined before and 1 hour after completion of SETs (Tables 1 and 2). In addition, plasma free fatty acids, serum BHBA, and plasma lactic acid concentrations were determined immediately after completion of SETs (Table 3). Because no significant differences were found in these measurements between control horses and IT horses, remaining results of this study are presented as mean ± SD values for all 10 horses, thereby eliminating the system of pairing horses for statistical analyses.

A significant (P = 0.003) effect of exercise was found on plasma 3-hydroxyisobutyric acid concentrations (Table 1), with consistently higher concentrations 60 minutes after every SET, compared with values before SETs. No significant effect of training and exercise was found on concentrations of other organic acids.

Exercise significantly decreased plasma C3-carnitine and C4-carnitine (P = 0.046 and P = 0.010, respectively; Table 3) concentrations, whereas plasma C2-carnitine concentrations significantly (P = 0.002) increased following exercise. Training did not affect plasma free fatty acids and acylcarnitine ester concentrations.

Plasma free fatty acid and lactic acid concentrations significantly (P = 0.004 and P < 0.001, respectively) increased immediately after exercise followed by a significant (P < 0.001) decrease 1 hour after exercise without reaching pre-exercise values (P = 0.020 and P < 0.001, respectively; Table 3). In comparison, serum BHBA concentrations significantly (P < 0.001) decreased immediately after exercise without reaching pre-exercise values at 1 hour after exercise (P = 0.009). No significant effect of training was found on plasma free fatty acid, serum BHBA, and lactic acid concentrations. During the phase 3 SET, IT horses maintained trotting at high speeds for a significantly (P = 0.012) shorter period of 16.1 ± 2.3 minutes, compared with 19.8 ± 0.4 minutes in control horses (equivalent to a 19% reduction).^p

Discussion

The right shift of the lactic acid concentration versus time curve during the phase 3 SET, compared with during the phase 2 SET, evident in both horse groups possibly indicates an adaptation to long-term training (Figure 1). However, the right shift of the lactic acid concentration versus time curve was still observed during the phase 4 SET for IT horses, but not for control horses, which might indicate that the IT horses adapted better to training than control horses as a result of the intensified training program during phase 3. Another explanation could be that the right shift is an indication of the lactic acid paradox, as described in people.^17,18

On the basis of values assessed prior to the phase 1 SET (Table 1), the predominant free fatty acids (C ≥ 14) measured in plasma were linoleic acid (C18:2), oleic acid (C18:1), palmitic acid (C16:0), and stearic acid (C18:0), presented in decreasing order. Together, these fatty acids constituted > 93% of the total concentration of C ≥ 14. Other detectable fatty acids, namely myristic acid and palmitoleic acid, constituted < 7% of the total concentration of C ≥ 14. These data correlate to those found in the study of Orme et al,¹⁹ where the predominant free fatty acids (C ≥ 14) in plasma were palmitic (C16:0), linoleic (C18:2), oleic (C18:1), stearic (C18:0), and linolenic acid (C18:3) in normally fed horses. Together, these fatty acids constituted > 90% of the total concentration of C ≥ 14. Other free fatty acids present were myristic (C14:0) and palmitoleic (C16:1), both of which constituted < 5% of the total concentration of C ≥ 14.¹⁹ In comparison, the major fatty acids in the horse carcass are as follows (in decreasing order): oleate, palmitate, and linoleate.²⁰

Immediately following the various SETs, lipolysis appeared to occur in horses of our study as reflected by the increased concentration of free fatty acids, taking into account that free fatty acid use is considered proportional to blood concentration (Table 3).²¹ Hambleton et al²² found that 4 long-chain fatty acids (palmitic, stearic, oleic, and linoleic) measured in plasma initially decreased with exercise, but following training, these long-chain fatty acids concentrations subsequently increased. Plasma palmitic, stearic, oleic, and linoleic acid concentrations in the current study were significantly affected by neither exercise nor by long-term training (Table 1). After deconditioning, however, analysis of plasma fatty acid concentrations during the phase 4 SET revealed an unexpected increase, albeit not a significant increase, in plasma fatty acid concentrations. Analysis of total free fatty acid revealed a significant increase in plasma concentration after exercise. The difference in changes in plasma concentrations between total free fatty acids and individual fatty acids may be attributable to the time of blood sample collection. Plasma total free fatty acid concentrations were measured from blood samples collected before, immediately after, and 1 hour after exercise. In contrast, plasma concentrations of individual fatty acids were measured from blood samples collected only before and 1 hour after exercise.

With 30 minutes of high-intensity exercise in humans, plasma short-chain and long-chain acylcarnitine concentrations increased by 46% and 23%, respectively.¹ In our study, mainly plasma short-chain acylcarnitine ester concentrations in Standardbreds were affected by exercise. Except for C3-carnitine, which can also be derived from the catabolism of amino acids methionine, valine, and isoleucine, these acyl groups are mainly degradation products of oxidation of long-chain fatty acids.²³ Results of our study revealed that plasma C2-carnitine concentrations significantly increased in response to exercise, whereas plasma concentrations of all other short-chain acylcarnitines decrease, which is in agreement with the increase in C2-carnitine content that occurs in the middle gluteal muscle in exercising Thoroughbreds.¹⁴

Training itself did not alter the concentrations of carnitine esters. Foster and Harris²⁴ described altered skeletal muscle carnitine concentrations in trained horses. In that study,²⁴ higher total skeletal muscle carnitine concentrations were found with increases in age and training as a consequence of underlying changes in mitochondria density as indicated by differences in citrate synthase activity. Foster et al²⁵ also reported higher plasma free carnitine concentrations of trained 3- to 6-year-old horses, compared with plasma concentrations of untrained age-matched horses. This is in contrast to results found in the present study, as the mean concentration of free carnitine in plasma was unaffected by training. It is not likely that the acclimatization to the treadmill exercise provided enough muscle conditioning to affect the results of our study because the adaptation period was not regarded as training per se. Rivero et al²⁶ found an additive effect of carnitine on muscular responses to training that should improve athletic performance. However, no alterations in plasma carnitine ester concentrations were found in our study. Hintz²⁷ casted doubt on the value of dietary supplementation of carnitine.²⁷

Measurements of BHBA provide a crude assessment of the rate of hepatic ketogenesis. It is 1 of the 3 ketone bodies, the other 2 being the volatile fatty acids acetone and acetoacetic acid, which are produced as a by-product of fatty acid metabolism.¹ Despite reaching the anaerobic threshold following the various SETs in our study, serum BHBA concentrations did not increase, which is in agreement with the fact that ketogenesis is an unlikely metabolic pathway in horses (Table 3).^28,29 On the other hand, exercise in our study was associated with an increase in the mean plasma concentration of the volatile fatty acid, 3-hydroxyisobutyric acid. This acid can be derived from the catabolism of valine, one of the branched chain amino acids, as a substrate of gluconeogenesis.³⁰

In animals, changes in the plasma and urine carnitine pools may reflect production from tissues other than muscle. Most commonly, enhanced ketogenesis is associated with an increase in plasma short-chain acylcarnitine concentration in humans and animals.⁴ No increase in serum BHBA concentration was detected in association with exercise-induced elevation in plasma C2-carnitine concentration in our study; thus, a hepatic source for the increase in plasma C2-carnitine concentration with exercise was unlikely in horses. Although short-chain (or volatile) fatty acids, mainly acetate (C2-carnitine), propionate (C3-carnitine), and butyrate (C4-carnitine), are produced via bacterial fermentation in the cecum and colon, increased absorption from the gastrointestinal tract during exercise seems unlikely because exercise leads to decreased perfusion and motility of the gastrointestinal tissues.³¹ As a consequence, the plasma acylcarnitine profile in horses might reflect skeletal muscle carnitine metabolism following exercise.

Acetyl coenzyme A is an immediate substrate of lipogenesis, and palmitic acid is the end product.³² When comparing plasma concentrations of palmitic acid (C16:0), for example, and other fatty acids in horses of our study, it appeared that carnitine formed esters predominantly with short- and medium-chain fatty acids.

Within an isolated training session, it is presently impossible to discern between short-term fatigue (termed overreaching) versus overtraining. This difficulty is partially the result of a lack of diagnostic tools, variability of results of research studies, a lack of wellcontrolled studies, and individual responses to training.³³ Overreaching is the short-term fatigue associated with the increases in exercise load that are usual in any training program.³⁴ In contrast, overtraining is a state of prolonged fatigue, caused primarily by an imbalance between training and recovery. Even with complete rest, recovery from overtraining can take weeks or months.³⁴ In our study, half of the horses were normally trained and the other half received intensified training with the goal of becoming overtrained.

Overtraining is defined as an accumulation of training and nontraining stress, resulting in long-term decrement in performance capacity with or without related physiologic and psychologic signs, of which restoration of performance capacity may take several weeks or months.³³ The reduction in performance seen in horses of our study is similar to that found in another study on horses,³⁴ in which the mean treadmill run-time to fatigue decreased by 14%. In our study, intensified training revealed no alterations in plasma free fatty acid and acylcarnitine ester concentrations. As a consequence, no particular marker for intensified training could be identified, and IT horses were unlikely to be associated with defective mitochondrial fat metabolism.

The relationship between skeletal muscle carnitine metabolism and recurrent exertional rhabdomyolysis needs further study. None of the horses in this study had a history of exertional rhabdomyolysis (serum creatine kinase activities were always within reference ranges, data not shown), but horses with rhabdomyolysis could well have alterations in muscle carnitine metabolism or other defects of metabolism, as Annandale et al³⁵ have documented an energy imbalance in the muscles of horses with recurrent rhabdomyolysis as a result of polysaccharide storage myopathy. Hence, there is a potential for carnitine metabolism abnormalities as a possible cause for energy deficit and rhabdomyolysis in horses with polysaccharide storage myopathy.