Secondary Logo

Journal Logo

Original Research

Small Beneficial Effect of Caffeinated Energy Drink Ingestion on Strength

Collier, Nora B.; Hardy, Michelle A.; Millard-Stafford, Mindy L.; Warren, Gordon L.

Author Information
The Journal of Strength & Conditioning Research: July 2016 - Volume 30 - Issue 7 - p 1862-1870
doi: 10.1519/JSC.0000000000001289
  • Free



Caffeine ingestion can acutely improve the ability to activate one's musculature during a maximal voluntary contraction, and as a consequence, strength and muscular endurance can be increased by as much as 7 and 18%, respectively (29). Products known as “energy” drinks share a common feature in that all contain caffeine and differ primarily from carbohydrate-electrolyte “sports drinks” based on this ingredient. Although caffeine intake for U.S. adults seems stable over time, energy drink consumption is markedly increasing (13) with reports that as many as half of all college athletes consume energy drinks (30). However, it is not clear if the ergogenic effect of caffeine on strength and muscular endurance exists (1), especially when caffeine is combined with a variety of other nutritional ingredients (e.g., carbohydrate, taurine, glucuronolactone, ginseng, and B vitamins).

The findings of studies investigating an energy drink's effect on strength are limited (23) and equivocal (3,4,11,27). Some studies have used formulated caffeinated products to mimic aspects of energy drinks but often fail to include ingredients commonly found in energy drinks (10). Using the commercially available energy drink, Red Bull (Red Bull GmbH, Salzburg, Austria), ingestion of 500 ml (2 mg·kg−1 body weight [BW] of caffeine) improved muscular endurance during bench press exercise; the number of repetitions that could be completed using a 70% 1 repetition maximum (1RM) load was increased by 6% (12). A study using a similar dosage of caffeine obtained from an energy drink also reported a significant 12% increase in the weight lifted during 4 sets of leg press but not bench press (3). However, Eckerson et al. (11) reported that sugar-free Red Bull (2 mg·kg−1 BW of caffeine) did not increase 1RM strength or the number of repetitions to failure in bench press. We recently assessed the effect of ingesting the energy drink, Full Throttle (The Coca-Cola Company, Atlanta, GA, USA), on hip adductor strength both before and during simulated high-performance aircraft flight (28). Strength measured under control (i.e., no flight) conditions was unaffected by energy drink ingestion but was improved by 29% during simulated flight compared with placebo. However, the improvement during flight was not different from that of an uncaffeinated version of the energy drink. The study by Del Coso et al. (9) provides the best evidence for a commercially available caffeinated energy drink improving strength compared with an uncaffeinated control drink. Isometric strength and isotonic power were found to be increased in both the half squat and bench press.

An important factor that may explain these mixed results is that most of the studies mentioned above used energy drinks delivering caffeine at a dosage lower than the range typically found to be ergogenic for improving strength or endurance exercise performance (i.e., 5–6 mg·kg−1 BW) (2,6,23,24,29). Furthermore, only 1 study (9) assessed an energy drink's effect on strength in a muscle group known to be most responsive to the effect of caffeine (i.e., the knee extensors) (29). It is also not clear from the studies observing an improvement in muscular strength and endurance whether the effect could be attributed to caffeine or some other component of the energy drinks. A systematic review on energy drinks concluded that additional well-designed, randomized, placebo-controlled studies are needed to assess efficacy and, if favorable, whether the effects of energy drinks are solely due to the caffeine content (22).

Therefore, the objectives of this study were to determine (a) if an energy drink providing a caffeine dose demonstrated to be ergogenic for muscular strength would improve maximum voluntary contraction strength in the knee extensors, (b) whether an observed improvement could be attributed to the caffeine in the energy drink, and (c) whether an improvement was due to an effect on the central nervous system (CNS) or peripherally on the muscle. Our hypothesis was that muscular strength and muscular endurance would be improved with the caffeinated energy drink (with carbohydrate and other ingredients) compared with placebo and would also provide a benefit over the uncaffeinated energy drink (containing only the carbohydrate and other ingredients).


Experimental Approach to the Problem

Our research questions were whether acute ingestion of a caffeinated energy drink improves muscular strength or muscular endurance, how it may or may not act to increase strength or muscular endurance, and if the potential effects can be attributed primarily to the effect of caffeine or its combination with other ingredients (e.g., carbohydrate) demonstrated to be ergogenic aids for exercise performance. This is of practical relevance because well-controlled, scientific evidence regarding the effect of energy drinks on strength is limited, and several of those studies have used caffeine dosages known to be less than optimal for improving exercise performance.

To answer this research question, a double-blind, placebo-controlled, repeated-measures experimental design was used. Thus, all subjects served as their own controls and were tested across 3 drink trials. One treatment was Full Throttle Original Citrus (The Coca-Cola Company), which contained carbonated water, high-fructose corn syrup, sucrose (carbohydrate totaling 29 g per 237 ml serving), taurine (amount not disclosed), natural and artificial flavors, citric acid, sodium benzoate, ginseng extract, caffeine (100 mg per 237 ml serving), guarana extract, acacia, carnitine fumarate, vitamin B3, vitamin B6, vitamin B12, glycerol ester of wood rosin, and yellow 5. The other 2 drink treatments were as follow: (a) Full Throttle without caffeine or guarana extract and (b) placebo (Full Throttle without any ingredients thought to be potentially ergogenic, i.e., caffeine, guarana extract, high-fructose corn syrup, sucrose, B vitamins, ginseng extract, carnitine fumarate, or taurine). All 3 drinks were made by the Coca-Cola Company, and each was stored in a container identified by a unique 3-digit code so that all researchers and subjects did not know the identity of the drinks. The codes were not broken until after the study had been completed and all data were analyzed.

After a preliminary testing session in which subjects were familiarized with all testing procedures, each subject performed 3 trials with 1 week between trials; each trial used a different one of the 3 drinks. All 3 trials for a subject began at the same time of day. The design goal was to balance the order of the drinks to avoid confounding caused by any potential carry over or learning effects. Although perfect balance is not possible with 15 subjects and 3 drinks, we created a partially balanced model in which each of the 6 drink-order permutations was used twice and 3 randomly selected permutations were used a third time. Subjects were randomly assigned to these drink orders.

On arrival in the laboratory for a trial, the subject rested in a chair for 10 minutes, which was followed by blood sampling from an antecubital vein to obtain plasma and serum. The subject was then instrumented for strength assessment of the right knee extensors (described below) with both stimulating and surface electromyographic (EMG) electrodes. The subject then performed a 10-minute warm-up on a cycle ergometer at a work rate eliciting a target heart rate of 61% of the person's estimated maximal heart rate (i.e., 220 − subject's age). This work rate was designed to elicit ∼50% of the person's maximal oxygen uptake. Five minutes after the warm-up, the strength assessment began.

After the strength assessment, the subject drank a volume of drink equal to 10.95 ml·kg−1 BW and rested for 2 hours. The subject then drank 5.48 ml·kg−1 BW of the drink. The total volume of drink was chosen because it provided 5 mg·kg−1 BW of caffeine when the caffeinated energy drink was consumed. The timing of the drink ingestions was chosen to match that of a companion study previously published (28) because we wanted to be able to compare outcomes. Thirty minutes after ingesting the second volume of drink, venous blood was sampled again. The cycling warm-up and strength assessments were then repeated. The subject then performed a fatiguing bout of isotonic concentric contractions using his or her right knee extensors (described in detail below). One minute after the bout of fatiguing contractions, the strength assessment was repeated for a third time. Finally, a venous blood sample was taken. The time line for a trial is depicted in Figure 1.

Figure 1
Figure 1:
Time line for an experimental trial. The times below the line indicate hours and minutes into the trial.


Potential subjects were recruited by word of mouth from a university student population. The potential subjects were screened using a health assessment form to ensure that they possessed no contraindications for participation. Contraindications included younger than 18 or older than 39, musculoskeletal injuries of the lower extremities within the last year, pregnant or trying to become pregnant, cardiopulmonary disease, disorders of the nervous system, systemic disease, and cancer. Potential subjects were excluded if they were nonusers of caffeine, had high daily caffeine intake (>6 mg·kg−1 BW), or were hypersensitive to caffeine. These factors were determined using a self-reported caffeine history questionnaire and a 7-day recall as previously published (8). Subjects also listed the use of any supplement in their health assessment form and those who consumed anything other than a multivitamin were excluded.

After meeting the above inclusion criteria, potential subjects also had to demonstrate the ability to perform the interpolated twitch procedure in a satisfactory and consistent manner (described below). Fifteen of 21 potential subjects were able to do so, and these subjects participated in and completed the study; there were 7 males and 8 females. All subjects were familiar with resistance training and physically active on a recreational basis but were not highly trained strength or endurance athletes who involved in a formal training program. The subjects' mean (±SD) physical characteristics were age 26.1 ± 3.5 years, height 174.6 ± 6.0 cm, and weight 70.7 ± 12.1 kg. Daily caffeine consumption averaged 142 ± 120 mg or 2.1 ± 1.8 mg·kg−1 BW. Before each test session, subjects refrained from eating or drinking anything but water during the 6 hours before testing and replicated their meals for the preceding 48 hours to standardize diet. Subjects also refrained from consuming any caffeine-containing food, beverage, or drug in the 14 hours before testing. The purpose of the caloric restraint period and standardization of diet was to ensure that the plasma glucose levels before the start of a trial were comparable for all 3 trials, whereas the caffeine restraint period was to ensure that the serum caffeine level was low (i.e., <5 μM) before the start of a trial. Both of these objectives were attained. The study's protocol was approved by the university's institutional review board, and all subjects gave written informed consent before participating in the study.


Strength Assessment

The interpolated twitch procedure was used to conduct the strength test and was conducted similarly to that described previously (8,25). The subject was seated in a KinCom III dynamometer (Chattecx, Chattanooga, TN, USA) in a semireclined position with 110° of hip flexion and 70° of knee flexion. These joint angles are those we have determined to yield the highest maximal voluntary isometric contraction (MVIC) and electrically evoked torques (EET) about the knee for the knee extensor muscle group using our instrumentation. The subject was positioned so that his or her knee's axis of rotation coincided with that of the servomotor. A seat belt along with chest, thigh, and ankle straps were used to hold the subject securely in place. To ensure consistent positioning of the subject from 1 test session to the next, the seat back position, seat position in the horizontal direction, lever arm length, and height of the servomotor axis of rotation were recorded during testing on the first trial. Those settings were used in all subsequent trials.

For stimulating the knee extensors, two 7- × 10-cm adhesive electrodes (UniPatch 620SS; Wabasha, MN, USA) were placed on the skin overlying the thigh, 1 over the distal vastus medialis muscle and the other over the proximal vastus lateralis muscle near the anterior superior iliac spine of the ilium (ASIS). Impedance of the electrodes was recorded using an impedance meter (Grass Technologies model EZM5D; Braintree, MA, USA) both before and after testing. Impedance was checked to ensure that the electrodes were not damaged and that there was adequate adhesion to the skin. Electrode impedance was required to be less than 10 kΩ. The electrodes were connected to a constant-current stimulator (Digitimer model DS7AH; Hertfordshire, England) that was controlled using a 667-MHz Pentium computer, a 16-bit A/D-interface and D/A-interface board (Keithley Instruments model KPCI 3108; Cleveland, OH, USA), and custom-written software created with TestPoint ver. 7.0 (Capital Equipment Co., Billerica, MA, USA). The software and interface board also sampled the torque output signal from the KinCom III dynamometer at 5,000 Hz.

To determine the stimulation current needed for subsequent procedures, a series of electrically stimulated isometric contractions of the knee extensors was performed with the current being increased on succeeding stimulations. Each stimulation consisted of a paired-pulse stimulation, i.e., two 0.2-millisecond pulses separated by 10 milliseconds. Stimulator current was initially set to 100 mA and was increased by 20 mA on each succeeding stimulation. The stimulations were delivered once every 20 seconds until peak contraction torque reached a plateau and then showed a decline on 2 successive stimulations. The current eliciting the highest peak torque on the plateau of the torque-current curve was used for the remainder of the test session. Determination of the optimum stimulation current was performed in the first 2 strength assessments of a trial but not in the third because of the necessity to start the interpolated twitch procedure 1 minute after the fatiguing bout of contractions. The current used in the third strength assessment was that determined to be optimum in the second assessment (i.e., that determined ∼25 minutes earlier). Optimum current usually changed no more than 20 mA between the first 2 assessments.

For the interpolated twitch procedure, the subject was instructed to perform a 3-second MVIC of his or her knee extensor muscles. Auditory cues elicited by the custom-written software were used to signal the subject to start and stop the contraction. At 2.5 seconds into the MVIC, the muscle group was stimulated with a paired-pulse stimulation, and the increase in torque over the MVIC level (i.e., interpolated twitch torque [ITT]) was measured. At 2 and 4 seconds after the end of the MVIC, the paired-pulse stimulation was administered to relaxed muscle to determine peak EET; the average value for the 2 stimulations was used in the data analyses. The percent muscle activation during MVIC was calculated as 100% × (1 − ITT/EET). Peak MVIC torque was defined as the average torque between 2 and 2.5 seconds into the MVIC. During a given test session, 6 interpolated twitch procedures were performed, with 1 minute of rest between procedures. Of the 6 interpolated twitch procedures, the 3 best attempts were determined and their data averaged together for use in subsequent analyses. Test-retest reliability was assessed for measurements of MVIC strength, EET, and percent activation during MVIC on 10 subjects on 2 tests with 7 days between tests. The intraclass correlation coefficient equaled or exceeded 0.84 for all measures.


Skin was prepared by shaving any hair and cleaning with an alcohol wipe. An active EMG electrode assembly (Delsys DE-2.1; Boston, MA, USA) was placed at 33% of the distance between the superior lateral patella and ASIS on the vastus lateralis muscle, whereas another electrode assembly was placed over the vastus medialis muscle at 30% of the distance between the superior medial patella and ASIS (26). The longitudinal axis of each electrode assembly was aligned parallel to the longitudinal axis of the underlying muscle fibers. An adhesive 2- × 2-cm reference electrode was placed on the skin overlying the fibular head. The electrodes were connected to an EMG amplifier (Delsys Bagnoli 8-channel EMG system) with a gain setting of 1,000 and a bandpass filter setting of 20–450 Hz. The amplifier was sampled at 5,000 Hz using the same computer and A/D board as used in the interpolated twitch procedure. A muscle's EMG root-mean-square (RMS) was calculated on its EMG signal between 2 and 2.5 seconds into the MVIC.

Fatiguing Bout of Isotonic Concentric Contractions

The subject performed 5 sets of 10 isotonic concentric contractions performed by the knee extensors on the KinCom III dynamometer. There were 5 seconds between contractions and 20 seconds between sets. The concentric movement was performed at 30°·s−1 over a 45° range beginning at 90° of knee flexion. The isotonic load was set at 55% of the peak MVIC torque determined that day before ingestion of any drink. The load was designed to elicit fatigue, defined as the subject being unable to complete without assistance, the tenth or earlier contraction in the second set. When a subject struggled to complete a contraction, a researcher provided the minimum amount of assistance necessary to ensure that the subject could complete the contraction, and that 50 repetitions were performed during the bout.

Blood Assays

To better understand the drinks' effects on strength, blood levels of caffeine, glucose, and insulin were assessed. Serum caffeine concentration was determined using a homogeneous enzyme immunoassay technique using commercially available reagents (Emit Caffeine Assay; Dade-Behring Syva, Cupertino, CA, USA). Plasma glucose concentration was assayed using a kit based on the hexokinase/glucose-6-phosphate dehydrogenase technique (Pointe Scientific catalog #G7517; Canton, MI, USA). Serum insulin concentration was assayed by enzyme-linked immunosorbent assay (Calbiotech catalog # IS130D; Spring Valley, CA, USA). All assays were run in duplicate. For the first 5 subjects, blood was only obtained at the first 2 time points. The caffeine, glucose, and insulin concentration data presented in the Results represent that of the last 10 subjects for which we obtained blood at all 3 time points.

Statistical Analyses

To assess the effect of drink type on a dependent variable, a 2-way (drink type × time) repeated-measures analysis of variance (ANOVA) was used. When a significant effect of drink type or an interaction was found, pairwise comparisons were made using single-degree-of-freedom contrasts with a Bonferroni correction. All statistical testing was conducted using SPSS (ver. 16) and an overall α level of 0.05. Values in the Results are reported as mean values (±SE). Using data from a previous similar study (8,25), we estimated the statistical power at 78% for the measurement of our noisiest dependent variable (i.e., percent muscle activation) while using 15 subjects and trying to detect a 7% change in the variable.


Strength Assessments

There was a significant main effect of drink type on the changes in MVIC strength (p = 0.017); there was no significant drink type-time interaction (p = 0.37). Ingestion of the caffeinated energy drink yielded changes in strength that were significantly different from those after ingestion of placebo (p = 0.015). Compared to predrink, caffeinated energy drink increased strength on average by 5.0% at 30 minutes after ingestion of the second drink; strength after placebo ingestion remained virtually unchanged (i.e., down by 0.5% on average) (Figure 2A). At this time point, the difference between the caffeinated energy drink and placebo equates to a Cohen's d effect size of 0.86 with a 95% confidence interval of 0.17–1.56. A Cohen's d of 0.86 is considered large according to Cohen (5). The difference between the caffeinated energy drink and placebo remained after the fatiguing bout of concentric contractions. The change in MVIC strength for the uncaffeinated energy drink was not significantly different (p ≥ 0.052) from that for either of the other 2 drinks at either of the 2 time points.

Figure 2
Figure 2:
Effect of the 3 drinks on maximum voluntary isometric contraction (MVIC strength). A) Mean (±SE) effects of the drinks. “Predrink” refers to the baseline measurement; “30 minutes after second drink” is the measurement taken before the fatiguing bout of contractions and “after exercise” is the measurement taken immediately after the fatiguing bout of concentric contractions. Asterisk denotes a significant difference (p ≤ 0.05) between the caffeinated energy drink and placebo at that time point. B) Individual responses from predrink to 30 minutes after the second drink for each of the 3 drinks. Each subject is represented by a different line/symbol. The heavy line/symbol represents the mean effect.

For all 3 drinks, the change in strength from predrink to 30 minutes after ingestion of the second drink is shown for each subject in Figure 2B. Visual inspection of these data indicates that 3–5 subjects might be classified as nonresponders to the caffeinated energy drink because the improvement in strength for these subjects to that drink was numerically the same or less than that for placebo. However, if the subjects could be categorized as either a responder or nonresponder to the effect of caffeine, one would predict that there would be a bimodal distribution in the individual caffeinated energy drink-placebo differences. There was no evidence of a nonnormal distribution, which would be needed to confirm a bimodal distribution. The Shapiro-Wilk test of normality for the individual differences was not statistically significant (p = 0.47).

The increase in MVIC strength after caffeinated energy drink ingestion is difficult to attribute to either a peripheral effect on muscle or an effect on the CNS. A peripheral effect is discounted because there was no significant drink type main effect or interaction on electrically stimulated strength (p ≥ 0.46) (Figure 3). A CNS effect is not believed likely because there was no significant drink type main effect or interaction on percent muscle activation during the MVIC (p ≥ 0.11) (Figure 4), or on vastus medialis or vastus lateralis EMG RMS (p ≥ 0.07) (Figure 5).

Figure 3
Figure 3:
Effect of the 3 drinks on electrically stimulated strength. There were no significant differences among drinks.
Figure 4
Figure 4:
Effect of the 3 drinks on percent muscle activation during maximum voluntary isometric contraction (MVIC). Mean activation at predrink was 89.2 (±1.1)%. There were no significant differences among drinks.
Figure 5
Figure 5:
Effect of the 3 drinks on electromyographic (EMG) root-mean-square (RMS) during maximum voluntary isometric contraction (MVIC). There were no significant differences among drinks for either muscle. A) Vastus medialis muscle, (B) Vastus lateralis muscle.

Blood Caffeine, Glucose, and Insulin Concentrations

Mean serum caffeine concentration was less than 4 μM at all times for the placebo and uncaffeinated energy drink conditions and also at the predrink time for the caffeinated energy drink condition, indicating subject compliance with test instructions. For the caffeinated energy drink, mean caffeine concentration had increased to 33.8 (±1.4) μM by 30 minutes after ingesting the second drink volume and to 36.9 (±2.4) μM after the fatiguing bout of concentric contractions.

Plasma glucose concentrations after drink ingestion are illustrated in Figure 6. There was a significant drink type-time interaction (p = 0.01). In the caffeinated energy drink trial, mean plasma glucose concentration at 30 minutes after ingestion of the second drink volume was significantly elevated compared with placebo (114 vs. 101 mg·kg−1). However, after fatiguing exercise, no difference was observed among the drink conditions.

Figure 6
Figure 6:
Effect of the 3 drinks on plasma glucose concentration. Asterisk denotes a significant difference (p ≤ 0.05) between the caffeinated energy drink and placebo at that time point.

In an attempt to explain the elevated glucose concentration for the caffeinated energy drink condition, serum insulin concentration was assayed. There was a significant drink type-time interaction (p = 0.01). At 30 minutes after ingesting the second drink volume, the caffeinated energy drink elicited a mean serum insulin concentration that was significantly greater than for the other 2 drinks (i.e., 42 vs. 11–31 μIU·ml−1) (Figure 7). However, after fatiguing exercise, no difference was observed among the drink conditions.

Figure 7
Figure 7:
Effect of the 3 drinks on serum insulin concentration. Asterisk denotes a significant difference (p ≤ 0.05) between the caffeinated energy drink and placebo at that time point. Dagger denotes a significant difference between the caffeinated and uncaffeinated energy drinks at that time point.


The study's findings indicate that when compared with placebo, ingestion of a caffeinated energy drink can acutely improve strength by 5%, and this advantage is maintained even after a fatiguing bout of exercise. The magnitude of our improvement is comparable with that Del Coso et al. (9) observed in their energy drink trial using 3 mg·kg−1 BW of caffeine. Extrapolating from their force-velocity curves, isometric strength can be estimated to have improved by 7 and 10% in the bench press and half squat, respectively. Likewise, our observed strength improvement is similar to the increase typically observed in knee extensor strength after ingestion of caffeine alone (i.e., 7%) as we reported in a recent systematic review and meta-analysis (29). The mechanism for the strength improvement observed in this study is unclear, i.e., whether the effect was manifested peripherally on the muscle or on the CNS. Finally, it is also not clear if the ergogenic effect of the energy drink can be attributed to caffeine or some other component of the drink. These latter 2 points are discussed below.

The mechanism for the energy drink-induced strength improvement we observed cannot be clearly explained by the study's data. The caffeinated energy drink appeared to have no effect on electrically stimulated strength, which if it had would indicate an effect peripherally on the muscle. Similarly, the energy drink did not seem to exert an effect on the CNS to improve strength because both percent muscle activation and EMG amplitude during MVIC were not significantly altered. Assuming that caffeine in the energy drink was responsible for the strength improvement, the most logical explanation for the improvement based on the literature is that percent muscle activation was increased by the energy drink. In our recent systematic review and meta-analysis investigating caffeine's effect on strength (29), we found convincing evidence for caffeine ingestion to increase percent muscle activation during MVIC (i.e., a moderate-to-large effect size of 0.67; p < 0.0001), whereas evidence for a direct effect of caffeine on muscle was weak (i.e., a trivial, nonsignificant effect size of 0.12). Our failure in this study to detect an energy drink effect on percent muscle activation can probably be attributed to inadequate statistical power. The expected drink effect on percent muscle activation should be small based on the relatively small effect observed on MVIC strength (i.e., ∼5% increase). Furthermore, the estimation of percent muscle activation is a relatively noisy technique compared with the measurement of MVIC and electrically stimulated strengths. Both of these factors would act to decrease statistical power. We should, however, point out that the trend in our data was for the caffeinated energy drink to yield the numerically highest percent muscle activations and vastus lateralis EMG amplitudes. Inadequate statistical power cannot be attributed to having both women and men as subjects. When gender was added as a factor to the 2-way ANOVA examining the drink effect on strength, there was no gender main effect or interaction (p ≥ 0.35). Studies with male subjects and those with a mixed-gender sample seem to respond similarly to caffeine (29). To our knowledge, no studies have tested the effect of caffeine on strength using only women.

This study casts doubt on the premise that caffeine is the sole ergogenic agent in the caffeinated energy drink. The best evidence for this is that there was no significant difference between the caffeinated and uncaffeinated energy drinks in MVIC strength change, either postingestion or after fatiguing exercise. This is an identical finding to that we made while investigating the energy drink's effect on MVIC strength during simulated high-performance aircraft flight (28). However, to confuse the issue, there was also no significant difference in this study between the uncaffeinated energy drink and placebo, particularly after fatigue. It is noteworthy that from predrink to 30 minutes after the second drink, the percent improvements in MVIC strength for the caffeinated and uncaffeinated energy drinks were numerically identical.

Of the ingredients in common for the 2 energy drinks but not in placebo, taurine and carbohydrate are the most logical alternatives to caffeine for explaining the improvement in strength. The evidence for taurine is not strong but includes the following. First, an observed improvement in high-intensity cycling exercise after ingestion of Red Bull has been partially attributed to taurine (14). Second, depletion and repletion of taurine from muscle decreases and increases, respectively, electrically stimulated force in rodents (15,19). McClellan and Lieberman (22) summarized that there is limited and inconsistent evidence that the addition of taurine to a caffeinated energy drink improves physical performance, based primarily on endurance tasks. We are not aware of any reports of acute taurine ingestion improving strength in animals or humans.

Carbohydrate ingestion before, during, and after endurance exercise has been studied extensively and in most instances shown to improve performance. However, there has been much less research into whether carbohydrate ingestion can improve strength. Reductions in muscle glycogen have been found to be associated with decreased strength (20). Carbohydrate ingestion increases the amount of work that can be performed during a resistance training session when muscle glycogen stores are low (7,18). We are unaware of any studies investigating how acute carbohydrate ingestion can affect strength when muscle glycogen stores are normal. Carbohydrate's effect in such a scenario may be dependent on the blood glucose level and whether the subject was fasting before the trial. Subjects in this study had fasted for at least 6 hours before the trial, and thus liver glycogen stores may have been low. Despite this, there was no evidence of hypoglycemia at the predrink time point; values for all 3 trials were ∼100 mg·dl−1 (Figure 6). If blood glucose level was related to MVIC strength, then one would not have expected identical strength improvements in the caffeinated and uncaffeinated energy drink trials at 30 minutes after the second drink volume because there were significantly disparate blood glucose levels (114 vs. 100 mg·dl−1). The relatively high blood glucose level at 30 minutes after the second drink volume in the caffeinated energy drink trial cannot be explained by a relatively low blood insulin level. In fact, the blood insulin level was significantly higher in the caffeinated energy drink trial than in either of the other 2 trials (Figure 7). This can be explained by a relative insulin insensitivity, which has been a consistent observation after caffeine ingestion (16,17). This difference in insulin levels, however, was abolished after fatiguing knee extensor exercise in this study. Carbohydrate may also act “centrally” since a recent study using carbohydrate mouth rinsing (swish and spit, not drinking) compared with placebo observed a small but significant improvement (Cohen's d effect size = 0.22) in countering fatigue-related strength reductions (21). To establish if carbohydrate (acting either centrally or “peripherally”) was responsible for our findings, a fourth trial using a sugar-free caffeinated energy drink would be needed.

From the above discussion, it would seem that caffeine and carbohydrate may both contribute to the strength-enhancing ergogenic property of the energy drink. However, the 2 ingredients may interact meaning that the ergogenicity of the 2 combined is less than the sum of the independent effects of the 2 ingredients. We have noted this before when caffeine is combined with carbohydrate, and the combination's effect on performance during endurance exercise is investigated. Caffeine seems to be less effective in improving endurance exercise performance when added to carbohydrate than when it is added to placebo or water (6).

In conclusion, we found that ingestion of a caffeinated energy drink can acutely improve strength by ∼5%, and this advantage is maintained even after a fatiguing bout of exercise. It is uncertain whether the strength improvement is due to an effect of the energy drink on the CNS or muscle tissue itself. Likewise, it is uncertain whether caffeine is the sole ingredient responsible for the strength gain; carbohydrate may also play a role.

Practical Applications

A caffeinated, sugar-containing energy drink consumed in the hours before training or competition improves strength and muscular endurance vs. a drink with no caffeine or sugar. Ingestion equivalent to approximately two 16-ounce cans (i.e., Full Throttle, Red Bull, or Monster) for a 60-kg person increases knee extensor muscle strength to a small extent and reduces fatigue during an exhaustive bout of high-force contractions. However, an uncaffeinated version of the drink gave the same level of performance, thus other ingredients (e.g., sugar) may be important in explaining the energy drink's ergogenicity. It is possible that a sugar-free version of energy drinks may not be as effective. Based on our findings and other prevailing published evidence, consumption of an energy drink may boost performance in strength-related activities, particularly during activities involving the knee extensors that are performed to exhaustion. Athletes should be advised that not all energy drinks are the same in caffeine content, with many delivering twice as much caffeine as the energy drink used in this study. Because some athletic-regulating agencies (e.g., National Collegiate Athletic Association [NCAA]) still restrict caffeine, college athletes should be advised to not consume excess caffeine, considering all dietary and supplemental sources, which could trigger a positive doping test during competition.


The authors thank Randal Glaser, Aaron Honeycutt, Britt Hubier, and Chris Ready for their assistance in the collection of the study's data. Sincere thanks to Dr. Maxime Buyckx and the Health and Wellness Programs, The Coca-Cola Company, Atlanta, GA for their assistance with the study.

This research used products provided by The Coca-Cola Company and was supported by a grant from The Coca-Cola Company. No other competing financial interests exist. The results of this study do not constitute endorsement of the product by the authors or the National Strength and Conditioning Association.


1. Astorino TA, Roberson DW. Efficacy of acute caffeine ingestion for short-term high-intensity exercise performance: A systematic review. J Strength Cond Res 24: 257–265, 2010.
2. Astorino TA, Terzi MN, Roberson DW, Burnett TR. Effect of two doses of caffeine on muscular function during isokinetic exercise. Med Sci Sports Exerc 42: 2205–2210, 2010.
3. Campbell BI, Downing J, Kilpatrick M, La Bounty P, Elkins A, Williams S, dos Santos MG, Chang T, Willey S, Kreider R. The effects of a commercially available energy drink on resistance training performance. Med Sci Sports Exerc 42: 448, 2010.
4. Carvajal-Sancho A, Moncada-Jiménez J. The acute effect of an energy drink on the physical and cognitive performance of male athletes. KinSI 11: 5–16, 2005.
5. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrence Erlbaum Associates, 1988.
6. Conger SA, Warren GL, Hardy MA, Millard-Stafford ML. Does caffeine added to carbohydrate provide additional ergogenic benefit for endurance? Int J Sport Nutr Exerc Metab 21: 71–84, 2011.
7. Conley MS, Stone MH. Carbohydrate ingestion/supplementation or resistance exercise and training. Sports Med 21: 7–17, 1996.
8. Cureton KJ, Warren GL, Millard-Stafford ML, Wingo JE, Trilk J, Buyckx M. Caffeinated sports drink: Ergogenic effects and possible mechanisms. Int J Sport Nutr Exerc Metab 17: 35–55, 2007.
9. Del Coso J, Salinero JJ, Gonzalez-Millan C, Abian-Vicen J, Perez-Gonzalez B. Dose response effects of a caffeine-containing energy drink on muscle performance: A repeated measures design. J Int Soc Sports Nutr 9: 21, 2012.
10. Duncan MJ, Smith M, Cook K, James RS. The acute effect of a caffeine-containing energy drink on mood state, readiness to invest effort, and resistance exercise to failure. J Strength Cond Res 26: 2858–2865, 2012.
11. Eckerson JM, Bull AJ, Baechle TR, Fischer CA, O'Brien DC, Moore GA, Yee JC, Pulverenti TS. Acute ingestion of sugar-free red bull energy drink has no effect on upper body strength and muscular endurance in resistance trained men. J Strength Cond Res 27: 2248–2254, 2013.
12. Forbes SC, Candow DG, Little JP, Magnus C, Chilibeck PD. Effect of Red Bull energy drink on repeated Wingate cycle performance and bench-press muscle endurance. Int J Sport Nutr Exerc Metab 17: 433–444, 2007.
13. Fulgoni VL, Keast DR, Lieberman HR. Trends in intake and sources of caffeine in the diets of US adults: 2001-2010. Am J Clin Nutr 101: 1081–1087, 2015.
14. Geiss KR, Jester I, Falke W, Hamm M, Waag KL. The Effect of a taurine-containing drink on performance in 10 endurance-athletes. Amino Acids 7: 45–56, 1994.
15. Goodman CA, Horvath D, Stathis C, Mori T, Croft K, Murphy RM, Hayes A. Taurine supplementation increases skeletal muscle force production and protects muscle function during and after high-frequency in vitro stimulation. J App Physiol 107: 144–154, 2009.
16. Graham TE, Sathasivam P, Rowland M, Marko N, Greer F, Battram D. Caffeine ingestion elevates plasma insulin response in humans during an oral glucose tolerance test. Can J Physiol Pharmacol 79: 559–565, 2001.
17. Greer F, Hudson R, Ross R, Graham T. Caffeine ingestion decreases glucose disposal during a hyperinsulinemic-euglycemic clamp in sedentary humans. Diabetes 50: 2349–2354, 2001.
18. Haff GG, Lehmkuhl MJ, McCoy LB, Stone MH. Carbohydrate supplementation and resistance training. J Strength Cond Res 17: 187–196, 2003.
19. Hamilton EJ, Berg HM, Easton CJ, Bakker AJ. The effect of taurine depletion on the contractile properties and fatigue in fast-twitch skeletal muscle of the mouse. Amino Acids 31: 273–278, 2006.
20. Jacobs I, Kaiser P, Tesch P. Muscle strength and fatigue after selective glycogen depletion in human skeletal muscle fibers. Eur J Appl Physiol 46: 47–53, 1981.
21. Jensen M, Stellingwerff T, Klimstra M. Carbohydrate mouthrinse counters fatigue related strength reduction. Int J Sport Nutr Exerc Metab 25: 252–261, 2015.
22. McLellan TM, Lieberman HR. Do energy drinks contain active components other than caffeine? Nutr Rev 70: 730–744, 2012.
23. Mora-Rodriguez R, Pallares JG. Performance outcomes and unwanted side effects associated with energy drinks. Nutr Rev 72(Suppl 1): 108–120, 2014.
24. Pallares JG, Fernandez-Elias VE, Ortega JF, Munoz G, Munoz-Guerra J, Mora-Rodriguez R. Neuromuscular responses to incremental caffeine doses: Performance and side effects. Med Sci Sports Exerc 45: 2184–2192, 2013.
25. Park ND, Maresca RD, McKibans KI, Morgan DR, Allen TS, Warren GL. Caffeine's enhancement of maximal voluntary strength and activation in uninjured but not injured muscle. Int J Sport Nutr Exerc Metab 18: 639–652, 2008.
26. Rainoldi A, Melchiorri G, Caruso I. A method for positioning electrodes during surface EMG recordings in lower limb muscles. J Neurosci Methods 134: 37–43, 2004.
27. Sünram-Lea SI, Owen-Lynch J, Robinson SJ, Jones E, Hu H. The effect of energy drinks on cortisol levels, cognition and mood during a fire-fighting exercise. Psychopharmacology (Berl) 219: 83–97, 2012.
28. Walker TB, Balldin U, Fischer J, Storm W, Warren GL. Acceleration tolerance after ingestion of a commercial energy drink. Aviat Space Environ Med 81: 1100–1106, 2010.
29. Warren GL, Park ND, Maresca RD, McKibans KI, Millard-Stafford ML. Effect of caffeine ingestion on muscular strength and endurance: A meta-analysis. Med Sci Sports Exerc 42: 1375–1387, 2010.
30. Woolsey C, Waigandt A, Beck NC. Athletes and energy drinks: Reported risk-taking and consequences from the combined use of alcohol and energy drinks. J Appl Sport Psychol 22: 65–71, 2010.

skeletal muscle; maximal voluntary contraction; interpolated twitch; carbohydrate

Copyright © 2016 by the National Strength & Conditioning Association.