Beta-Alanine Supplementation for CrossFit^® Performance

Abstract

This study aimed to investigate whether beta-alanine supplementation (BA) improves performance and rating of perceived exertion (RPE) and reduces the respiratory exchange ratio (RER) during a CrossFit^® workout. Fourteen participants were randomized in a double-blind design to either BA or placebo, with 12 participants (7 males, 5 females, 32 ± 9.2 y) completing the study. Participants performed two tests, separated by three weeks of supplementing with either 6.4 g/day of BA or placebo. Performance tests involved time to complete an adapted CrossFit^® “Fran” Workout of the Day: 21-15-9 repetition scheme alternating between dumbbell thrusters and kipping pull-ups. No significant differences between the BA group and the placebo group were observed for performance time improvement (−13.4 s vs. −12.9 s, p = 0.97), change in mean RER (0.06 vs. 0.05, p = 0.84), or change in RPE (10-point scale) (−0.4 vs. −0.07, p = 0.56). There was a group × time × time during test interaction for RER (p = 0.021). Compared to pre-testing, post-testing RER was higher at the 25% time point of the test for the BA group and at the 75% and 100% time points in the placebo group (p < 0.05). Beta-alanine did not show significant ergogenic effects during an adapted version of the CrossFit^® workout “Fran”, although it might have helped with the buffering of acidity later in the test, based on RER.

1. Introduction

CrossFit^® is a form of exercise that sometimes combines weightlifting, gymnastics, and aerobic movements in what are referred to as Workouts of the Day (WOD) in specific gym settings. Most WODs are performed at a high intensity, often demanding maximal effort, and eliciting high cardiovascular responses [1,2,3]. “Fran” is a WOD that consists of a 21-15-9 repetitions scheme alternating between two movements, “thrusters” (a combination of a barbell front squat into a shoulder press) and pull-ups (often performed in a kipping or butterfly variation) [4]. It is known for its high anaerobic demand and elicits respiratory exchange ratios greater than 1, elevated lactate levels, and high ratings of perceived exertion [5]. Although the high anaerobic demand for CrossFit^® workouts is known, limited research exists on supplements that could enhance CrossFit^® performance [6].

Beta-alanine (BA) is the rate-limiting substrate for carnosine synthesis, a dipeptide involved in intracellular acid-base regulation [7]. Four weeks of BA supplementation can increase skeletal muscle carnosine levels by nearly 60% [8,9]. Carnosine’s buffering capacity results from histidine, with the other amino acid involved in the synthesis, which has an imidazole functional group. The equilibrium between carnosine’s acid and base forms acts as a buffer against the pH decrease caused by acidic metabolic by-products during intense exercise in skeletal muscles [10].

BA supplementation has significant positive effects on exercise capacity and performance [7], likely due to its role in increasing carnosine buffering capacity during exercise [10,11]. However, most studies are limited to cardiovascular exercises such as cycling [9,12], swimming [13], or rowing [14,15]. There is no research investigating the effects of BA on CrossFit^® style workouts, which combine both resistance and aerobic elements.

Our purpose was to investigate the effects of BA supplementation on CrossFit^® performance. Given previous research supporting BA as an ergogenic buffering aid and the high anaerobic demands of many CrossFit^® WODs, we hypothesized that BA would improve performance during an adapted version of the “Fran” WOD. Our secondary purpose was to investigate the respiratory exchange ratio (RER) given that BA’s influence on acid-base regulation may contribute to variability in CO₂ production [16]. Specifically, we hypothesized that BA supplementation would result in a lower RER during the post-intervention test, despite similar or increased exercise intensity.

2. Methods

2.1. Participants

Participants (≥18 years of age) who were injury-free, had at least one year of CrossFit^® training experience (minimum frequency of 2 sessions per week), and were cleared by the Get Active Questionnaire (GAQ) to engage in physical activity were eligible to participate. This questionnaire screens for cardiovascular disease or chest pain, high blood pressure (with a cutoff of 160/90 mmHg), dizziness/lightheadedness during physical activity, shortness of breath at rest, loss of consciousness/fainting, concussion, acute injury, arthritis, back pain, diabetes, cancer, osteoporosis, asthma, and spinal cord injury. Participants needed to answer “no” to each of these to be included. Participants were excluded if they had started using performance-enhancing supplements within the past month or had regularly supplemented with BA within the past six months [17]. The use of other performance-enhancing supplements such as caffeine, creatine, branched-chain amino acids (BCAAs), vitamins, protein powder, or herbal products, taken regularly for more than one month were not considered exclusion criteria, but participants were told to maintain a stable dose of these supplements during the study. All participants provided written consent prior to participating. The study was approved by the Research Ethics Committee of the College of Kinesiology at the University of Saskatchewan, and study procedures were in accordance with the ethical standards of the Helsinki Declaration. The study was registered at clinicaltrials.gov (NCT05772988).

2.2. Protocol

The study involved two testing sessions separated by a three-week intervention in a randomized double-blind placebo-controlled manner. Participants were stratified by sex and randomly assigned to either the BA or placebo (PLA) group in a 1:1 ratio in blocks of four using an online random number generator (https://www.calculator.net/random-number-generator.html (accessed on 20 March 2023)). The BA group supplemented with 6.4 g of BA (Canadian Protein beta-alanine powder, Canada) mixed with 6.4 g of maltodextrin (Globe Plus 10 DE Maltodextrin, Univar Canada) per day for 21 days, and the PLA group supplemented with 12.8 g of maltodextrin per day for 21 days. Both groups divided their daily 12.8 g dose into four separate doses of 3.2 g, with a minimum two-hour gap between each dose. This dose of BA was chosen because significant exercise improvements are typically observed with doses ranging from 3.2 to 6.4 g/day [7,17]. Participants were required to maintain a daily log to track adherence and report any side effects.

Participants filled out the Get Active Questionnaire (GAQ) to screen for any contraindications to physical activity prior to arriving at their first testing session. At the start of each testing session, prior to engaging in any activity, participants’ height and weight were measured using a stadiometer and a portable ground scale, respectively.

An adapted version of the CrossFit^® workout “Fran” was performed during the testing sessions, once at baseline and then again following the three-week intervention. During the test, participants wore a COSMED K5 (COSMED, Rome, Italy) portable metabolic system to measure RER. The system was calibrated at the beginning of each test day according to the instructions provided by the manufacturer: Before each testing session, the COSMED K5 was calibrated for flow using a 3 L syringe and calibrated against known gas concentrations for oxygen and carbon dioxide. The respiratory exchange ratio was assessed throughout the entire workout without any demarcation for exercise phases. Participants were instructed to keep track of their dietary intake, activity level, sleep, and caffeine intake in the 48 h leading up to the baseline test and replicate these conditions before the post-intervention test. For 24 h prior to both tests, participants refrained from drugs, alcohol, and vigorous activity. They were also instructed to keep their diet, training, and current supplement regime consistent throughout the study.

Fran is a classic CrossFit^® benchmark workout of the day (WOD) consisting of a 21-15-9 repetition scheme alternating between thrusters (front squat into overhead press) and pull-ups. In this study, an adapted version of “Fran” was used, replacing barbell thrusters with dumbbell thrusters (2 × 25 lb for females or 2 × 40 lb for males) and kipping pull-ups (a modified pull-up technique) to accommodate for the COSMED portable metabolic system face mask. For the Fran workout testing, participants completed 21 dumbbell thrusters, followed by 21 pull-ups, then 15 thrusters, 15 pull-ups, and finally 9 thrusters and 9 pull-ups, completing each movement before moving on to the next. The repetitions within each set could be broken into smaller sets; however, the intent of the test was to finish as quickly as possible. The participant’s score was the time taken to complete all repetitions, recorded immediately following the last repetition of the set of 9 pull-ups [1]. Testing took place under the researcher’s supervision who ensured movement standards were met. Verbal encouragement was not used during the test, nor were spectators present. Participants were unaware of their baseline test scores during the post-intervention test, and the clock was hidden from the participants throughout the protocol to minimize external pacing strategies.

A standardized warm-up, adapted from Leitão et al. [4], was used during both testing sessions to prevent injury and familiarize participants with the portable metabolic system. The warm-up consisted of two parts. The first half was performed without the metabolic system and consisted of three minutes of rowing at a self-determined easy–moderate pace followed by five minutes of dynamic stretching. For the second half, participants wore the metabolic system and performed three sets of five to ten repetitions of specific movements, each progressing in difficulty until the movement standards were met in the final set. After the warm-up, participants rested approximately two minutes before starting the test. Immediately after each test, participants assessed their rating of perceived exertion using Borg’s 10-Point Rating of Perceived Exertion (RPE) scale [18]. The study design is summarized in Figure 1.

To better describe the fitness levels of participants, it was decided to measure aerobic capacity via a progressive treadmill test at the end of the study. Seven of the twelve participants who completed the original study agreed to this further testing and signed an amended consent form, approved by our university’s Research Ethics Board. The progressive-intensity treadmill testing was similar to that described elsewhere [1] except that we used a different metabolic cart (Sensor Medics, Vmax Series 29, Anaheim, CA, USA) for the assessment of breath-by-breath oxygen consumption.

2.3. Statistical Analysis

Performance was measured using time to complete all repetitions of the testing protocol and was assessed with a group (2) × time (2) ANOVA, with repeated measures on the “time” factor. Respiratory exchange ratio (RER) was measured at 20 s intervals. Mean RER was calculated based on the average of all RER values throughout the test. Moreover, 25% and 75% RER values were based on the average of the RER corresponding to the respective % of the participant’s pre-test time, along with the 20 s interval RER before and after it. End RER was based on the average of the last two RER values of the test. Mean RER, RPE, and dietary variables were assessed with a group (2) × time (2) ANOVA with repeated measures on the “time” factor. RER during the session was assessed by a group (2) × time during the session (3; i.e., 25%, 75%, and end-test) × time during the intervention (2; pre- vs. post-intervention) ANOVA with repeated measures on the last two factors. If there was a significant interaction from this ANOVA, an LSD post-hoc test was used to determine differences between pairs of means. All values are presented as means and SD. Statistics were evaluated using Statistica 7.0 (Statsoft, Chicago, IL, USA). Significance was accepted at p ≤ 0.05.

3. Results (See Data Available as Supplementary Material)

Participant flow through the study is shown in Figure 2. Fourteen participants were initially enrolled in the study (N = 7 BA; N = 7 PLA). However, 12 (N = 5 BA; N = 7 PLA; five females, seven males) participants completed the entire protocol and were included in the data analysis. Two participants did not complete the post-intervention test, one due to a shoulder injury unrelated to the testing protocol, and the other reported negative side effects (facial acne) considered to be “likely” associated with the BA intervention. Characteristics of participants who completed the study are outlined in

Table 1. Participant characteristics. Participant age, weight, height, years of CrossFit^®, and maximal aerobic capacity presented as Mean ± SD.

Variable	Group
	Beta-Alanine	Placebo
Sex	2 Female 3 Male	3 Female 4 Male
Age (years)	30 ± 9.8	34 ± 9.2
Weight (kg)	79.3 ± 14.0	73.8 ± 13.4
Height (cm)	173.0 ± 9.6	171.6 ± 7.7
Body mass Index (kg/m2)	22.8 ± 3.0	21.4 ± 3.1
Years of CrossFit®®	4.3 ± 3.4	8.4 ± 3.7
VO2peak (mL/kg/min) (n = 1 female and 3 males from the beta-alanine group and 3 males from the placebo group)	60 ± 13	56 ± 3
Number of Participants with Previous Experience doing the “Fran” Workout of the Day	4/5	7/7

Note: There were no differences between groups at baseline.

. There were no differences between groups for any of these characteristics (p > 0.05).

3.1. Performance

There was no significant difference between groups for change in performance (group × time; p = 0.97). The time to complete all repetitions of an adapted “Fran” WOD decreased for the BA group from 298 ± 21.2 s to 284.6 ± 20.7 s (−13.4 ± 27.6 s) compared to the PLA group, which decreased from 359.3 ± 89.1 s to 346.4 ± 95.9 s (−12.9 ± 26.4 s).

3.2. Respiratory Exchange Ratio

Due to a malfunction with the COSMED K5 portable metabolic system, data from four participants (2 BA; 2 PLA) were excluded from the analysis. There was no group x time interaction for mean RER (p = 0.84). There were no significant changes in mean RER from pre-test to post-test for the BA group (1.20 ± 0.08 vs. 1.27 ± 0.07) or the PLA group (1.16 ± 0.08 vs. 1.21 ± 0.06).

There was a significant “group” × “time during the session” × “time during the intervention” interaction for RER (p = 0.021), outlined in Figure 3. Post-hoc tests indicated RER was significantly higher in the BA group during the post-test compared to the pre-test at the 25% time point (p < 0.01), but not at the 75% time point (p = 0.76) or the end of the test (p = 0.32). RER was significantly higher in the PLA group during the post-test compared to the pre-test at the 75% time point (p < 0.01) and the end of the test (p = 0.011), but not at the 25% time point (p = 0.35).

3.3. Rating of Perceived Exertion

There was no group × time interaction for the rating of perceived exertion (p = 0.59) with the BA group changing from 7.5 ± 1.1 to. 7.1 ± 1.7 and the PLA group changing from 8.1 ± 0.5 to 8.0 ± 0.8.

3.4. Dietary Variables

Table 2. Mean ± SD dietary intake 48 h before the pre-intervention testing session and 48 h before the post-intervention testing session.

	Pre-Intervention				Post-Intervention
	Energy (kcal)	Protein (g)	Carbohydrate (g)	Fat (g)	Energy (kcal)	Protein (g)	Carbohydrate (g)	Fat (g)
Beta-alanine	1931 ± 324	115 ± 39	199 ± 63	78 ± 11	1888 ± 81	130 ± 11	197 ± 15	68 ± 23
Placebo	1955 ± 322	130 ± 51	215 ± 27	66 ± 9	1954 ± 420	135 ± 42	203 ± 44	68 ± 22

Note: There were no differences between groups over time.

shows changes in dietary variables (mean intake 48 h before the testing sessions). There was no difference between groups or over time for dietary variables.

4. Discussion

The current study is the first to assess the effects of chronic BA supplementation for CrossFit^® performance by combining thrusters and pullups in a Fran WOD. BA supplementation did not appear to improve performance during an adapted version of the CrossFit^® workout “Fran” nor reduce mean RER, although it might have helped with buffering acidity later in the test, based on RER.

4.1. Performance

Contrary to Saunders et al. [7] suggesting BA supplementation can improve exercise performance during bouts lasting between 0.5 and 10 min, BA did not appear to elicit any meaningful improvements in performance during an adapted version of the CrossFit^® WOD “Fran”. This disparity may be attributed to the testing protocol used, which is a performance test requiring participants to perform a set amount of work, rather than a capacity test that requires exertion until exhaustion [19]. Performance tests allow for internal pacing strategies that may not elicit maximal exertion, thus reducing the influence of carnosine buffering [19,20]. BA supplementation is almost twice as effective at improving exercise capacity (i.e., time to exhaustion) rather than performance [7], which may explain the lack of significant difference in the current study. Future studies evaluating the effects of BA could focus on events that are limited by time to exhaustion (i.e., total exercise capacity) rather than performance, which might be subject to pacing.

Although a dose of 1.6 g of BA for just two weeks has been shown to increase muscle carnosine stores, significant exercise improvements are typically observed with doses ranging from 3.2 to 6.4 g/day for 4–12 weeks [7,17]. In the current study, a three-week intervention length was chosen to promote participant adherence, reduce attrition rates, and investigate the effects of a shorter intervention. However, it is possible that three weeks was not sufficient for BA to produce notable ergogenic effects. This explanation is supported by Smith et al. [21] who investigated BA over a six-week period (6 g/day for the first three weeks then 3 g per day for the following three weeks) and observed that increases in peak O₂ utilization, time to fatigue, and total work carried out at 110% of the pre-training VO₂peak were only significant for the BA group after the second half of the intervention with no significant ergogenic effects evident in the initial three weeks.

Neither the BA group nor the PLA group showed a significant reduction in time during the post-intervention test, indicating no significant learning effect on our testing protocol. This contrasts with a study by Stein et al. [22], which found a significant learning effect during the CrossFit^® benchmark WOD “Cindy” and suggested the likelihood of learning effects in CrossFit^® workouts. In the current study, participants performed variations of the testing movements wearing the portable metabolic system as part of a standardized warm-up. This familiarization with the portable metabolic system and previous experience performing the “Fran” WOD in most of our participants (Table 1) would have attenuated any learning effect. Our participants were very experienced with CrossFit^® workouts, having a mean of 6.7 years of CrossFit^® experience with 11 of the 12 participants who completed the study having previous experience with the “Fran” WOD.

4.2. Respiratory Exchange Ratio

In the current study, high RER throughout the Fran workout indicated participants were working at very high intensities. At high exercise intensities, carbon dioxide output increases out of proportion to oxygen consumption in conjunction with increased lactic acid appearance [2]. This causes an increase in RER, with values greater than 1.1 (as seen during the Fran workout in the current study; Figure 3) indicating that participants were working at intensities close to their peak aerobic capacity [1]. Dietary variables were unlikely to influence RER as participants consumed the same amount and type of food before both tests [23]. An increase in acidity is associated with an increase in CO₂ without a subsequent increase in O₂, resulting in an increase in RER during intense activity [2]. Carnosine, which is increased with BA, is an intracellular buffer [16], and would theoretically decrease H+ ions available to be buffered extracellularly by the bicarbonate buffer system; however, this was not reflected in our observed changes in mean RER between groups. Three weeks may be insufficient for BA to significantly impact buffering capacity [21].

Although the mean RER was not significantly affected by BA supplementation, we observed differences within the BA and PLA groups during the Fran workout. In the BA group, RER was significantly higher at the 25% time point of the post-test but not at the 75% time point or the end of the test. The elevated RER during the early stages of the test was unexpected; however, the subsequent decrease in RER to pre-test levels can potentially be attributed to BA’s role in intracellular buffering [10]. Relative to the 25% time point, this would decrease blood acidity, potentially affecting CO₂ output relative to oxygen consumption and decreasing RER. In the PLA group, RER was significantly higher at the 75% and end-of-test time point of the post-test but not at the 25% time point. These increases during the post-test observed within the BA group and PLA group may be explained by variations in the movement performed at the time of gas analysis. Differences in posture as well as skeletal muscle recruitment can elicit different metabolic responses [24]. Therefore, if a participant was performing a thruster at a certain time point during the pre-test but a kipping pull-up at that same time point during the post-test, or vice versa, their metabolic response throughout the test may appear different at that time point.

The RER values observed in the current study, while relatively high, echo the current literature on the metabolic demands of CrossFit^® workouts. In a study by Fernández-Fernández et al. [5], participants spent 76% of the time during “Fran” with an RER > 1. The high-intensity nature of CrossFit^® workouts is further supported by Kliszczewicz et al. [25], who found that CrossFit^® workouts can elicit a metabolic response of 9.5 metabolic equivalents (METs).

4.3. Rating of Perceived Exertion

On average, participants reported high ratings of perceived exertion during the pre-test (7.9 ± 0.8) and post-test (7.7 ± 1.3) based on Borg’s 10-point RPE scale [18]. These values align with Fernández-Fernández et al. [5], who observed that “Fran” elicited an RPE of ~8.4, reflecting the high-intensity nature of the workout. The similar change in RPE between groups over time is contrary to our hypothesis that BA supplementation would make the workout seem easier; however, our results are in alignment with Huerta Ojeda et al. [25], who found no significant differences in RPE with BA supplementation. As the level of demand did not change across tests, RPE likely remained consistent because the participants consistently exerted themselves at the upper limit of the RPE scale throughout both tests [25].

4.4. Strengths

Our study had several strengths. The study utilized a randomized, double-blind design with a placebo-controlled group. Testing conditions were made consistent across both tests by using a pre-test diet and activity logs, and by conducting both intervention tests at the same time of day.

4.5. Limitations

The primary limitations of this study were the small sample size, which limited statistical power, and the intervention length, which may have been insufficient to produce meaningful changes in performance or RER. It was difficult to determine an apriori sample size because there are no previous studies using BA supplementation during CrossFit^® exercises. Other studies that have evaluated a similar dose and duration of BA supplementation as used in the current study resulted in an improvement in supramaximal cycling time to exhaustion of 15.2 s compared to −2.4 s for a placebo group, with a standard deviation of 16.6 s [26]. Using an alpha of 0.05 and power of 80%, this indicates a required sample size of 15 per group, indicating we may have been underpowered in our study. A p-value of 0.05 was used for all statistical testing. There is a possibility of inflation of type I error with multiple comparisons in our study. In our study, we included participants who had been taking stable doses of other nutritional supplements for at least one month and this may have influenced results. We decided it would not be feasible to recruit CrossFit^® athletes by excluding athletes taking any nutritional supplement because the majority of CrossFit^® athletes (>80%) regularly take nutritional supplements [27]. Our study duration of 3 weeks may have been insufficient to observe the beneficial effects of BA: Previous studies have indicated beneficial effects using similar doses of BA as used in the current study in 4–12 weeks [7,17]. We chose a 3-week duration mainly due to participants’ availability during the university semester. Finally, 50% of participants in the BA group reported paresthesia, a known side effect of BA [28], potentially affecting blinding and perhaps negatively affecting performance. A suggestion for future research is to assess a large sample size of CrossFit^® athletes and to also evaluate the effectiveness of BA during different CrossFit^® routines.

5. Conclusions

Three weeks of BA supplementation was not effective for enhancing CrossFit^® workout performance and had minimal effect on RPE and RER.