During high-intensity cycling time trials (TT), such as a 4-km TT, the distribution of power output (PO) throughout the trial (i.e., pacing) and, consequently, overall performance, seems to be preestablished by an anticipatory response based on previous experience and estimated exercise end point (9,11,32). However, during the trial, a copy of the neural drive (corollary discharge) may provide a signal to sensory areas within the cortex (feedforward mechanism), which could influence the effort perception and ultimately promote adjustment in PO (26). In addition, afferent signals from peripheral tissues to the sensory areas within the cortex also contribute to regulate PO, preventing early critical metabolic disturbances as an early depletion of energy stores and/or an early build up of metabolites (2,9). Indeed, the blocking of afferent feedback from contracting locomotor muscles by interspinous ligament injection of intrathecal fentanyl results in an overshoot of the initial PO chosen by participants, which results in a greater rate of accumulation of muscle metabolites and increased peripheral muscle fatigue development (2). The coordinate action of both feedforward and feedback mechanisms has been recently termed “sensory tolerance limit” and may explain exercise limitation in more global terms (26). The sensory tolerance limit accounts for the sum of all feedback (mainly from locomotor and respiratory muscles) and feedforward signals processed within the CNS with the purpose of regulating the intensity of exercise to ensure that voluntary activity remains tolerable.
Within several potential afferent signals that could affect sensory tolerance limit and ultimately alter exercise performance, muscle and/or blood pH might play a critical role during high-intensity exercise such as 4-km cycling TT (20). Anaerobic metabolism is the main factor affecting muscle and blood pH during exercise, and it is interesting to note that the anaerobic energy expenditure rate mirrors the PO profile during a 4-km TT (11,19). It is also noteworthy that the amount of anaerobic energy that can be produced during a 4-km cycling TT is assumed to be a constant value (9,19). These observations suggest that anaerobic metabolism during a 4-km cycling TT must be strictly regulated and this might be made by central monitoring of pH changes (17,18). Although the exact mechanism is not known, an elevated anaerobic metabolism during intense exercise leads to an accumulation of H+ in the muscle and blood, which may increase afferent feedback signals to the CNS. Acting in a contraregulatory way, descending drive from the primary motor cortex may be reduced and thereby reduce PO and ultimately the use of anaerobic system (32). This negative feedback loop might prevent an excessive and premature accumulation of H+ that otherwise would lead not only to premature fatigue but also to a premature reduction of the capacity-limited anaerobic reserve (20).
Accordingly, if this hypothesis is correct, then it would be expected that inducing a low blood pH (metabolic acidosis) may result in a concomitant reduction of PO and anaerobic energy expenditure rate during a 4-km cycling TT. On the other hand, inducing a high blood pH (metabolic alkalosis) may result in an increase in PO and anaerobic energy expenditure rate. A further prediction of this model is that the amount of anaerobic energy produced during the trial should reduce in metabolic acidosis and increase in metabolic alkalosis. Although the first assumption that metabolic acidosis would reduce the amount of anaerobic energy produced seems to be plausible, there is no experimental data to prove it. In addition, data showing the effect of metabolic alkalosis on anaerobic metabolism are controversial. There is evidence to suggest that preexercise-induced alkalosis increases muscle lactate concentration after short-duration cycle sprints (4), indicating an increased anaerobic energy contribution. On the other hand, the total amount of work done above critical power (assumed as an indicator of anaerobic capacity) does not alter with preexercise-induced alkalosis (39). It should be noted, however, that the total amount of work done above critical power has been largely criticized as a parameter strictly representing a fixed anaerobic work capacity (for a review, see Poole et al. ). Therefore, the effect of preexercise acidosis and alkalosis on anaerobic energy expenditure rate and the amount of anaerobic energy produced during a 4-km cycling TT is an unresolved question.
Acidosis and alkalosis during exercise have been manipulated using the oral intake of agents such as ammonium chloride (NH4Cl) and sodium bicarbonate (NaHCO3), respectively (4,8,14,29,36). The ingestion of 0.3 g·kg−1 body mass (BM) of NH4Cl is associated with a decrease in the time to exhaustion during intense constant-load exercise (~90%–110% V˙O2max) lasting approximately 3–5 min (28,36). This NH4Cl-induced decrement in exercise performance has been attributed to a reduction in the H+ efflux from the muscle to the blood during the exercise, which promotes intracellular H+ accumulation. An important point concerning the NH4Cl-induced effect on performance is that the typical 0.3 g·kg−1 BM dose is reported to be associated with gastrointestinal discomfort, which per se could affect performance (25,37). By contrast, acute ingestion of 0.3 g·kg−1 BM of NaHCO3 has been shown to be effective in improving performance during intense constant-load exercise performed until exhaustion (~90%–100% V˙O2max) lasting approximately 5–10 min (21,28,35). This improvement in exercise performance after NaHCO3 ingestion is associated with an increased H+ efflux from muscle to the blood, which attenuates the exercise-induced decline in intracellular pH (3,35). However, such time-to-exhaustion type tests do not permit athletes to adjust their external PO, so the effect of manipulating blood pH on regulation of the PO and anaerobic energy expenditure rate is not able to be elucidated from such studies.
Therefore, we investigated whether changes in blood pH, through preexercise ingestion of acidifying and alkalizing agents, affect PO, anaerobic energy expenditure rate, and performance during a 4-km TT. We hypothesized that the NH4Cl-induced reduction in blood pH would result in a reduced PO and anaerobic energy expenditure rate throughout 4-km TT, which would reduce the amount of anaerobic energy produced and overall performance, whereas the NaHCO3 ingestion would have the opposite effect. Because higher doses of NH4Cl are related to strong sensations of gastrointestinal discomfort (25,37), and an optimal ingestion protocol has not been fully established, we conducted a preliminary study with a subsample investigating whether a reduced and fractioned dose of NH4Cl would cause metabolic acidosis without inducing substantial gastrointestinal discomfort. This was important to examine NH4Cl effects independent of extraneous factors.
Seven healthy men (age = 26.1 ± 2.3 yr, mass = 70.2 ± 6.2 kg, height = 1.76 ± 0.10 m) volunteered to test the optimal NH4Cl, NaHCO3, and calcium carbonate (CaCO3, placebo) doses to be ingested in the main study with minimal gastrointestinal disturbance. All study procedures, as well as the protocol, benefits and risks were explained before the first visit, and participants gave their written informed consent. This study was approved by the Ethics Committee of the University of São Paulo.
Participants of this preliminary study attended the laboratory on four different occasions. During the first visit, participants underwent measures of BM and height, completed a dietary record of all food consumed (type, amount, and time) 24 h before their arrival at the laboratory, and received instructions for the next sessions. On the three subsequent visits, using a double-blind, counterbalanced, and repeated-measures design, the participants ingested NH4Cl, NaHCO3, or calcium carbonate (CaCO3, placebo).
Participants followed the same diet 24 h before and consumed their last meal 2 h before each experimental session (the same meals registered in the food recall obtained at the first visit). On arrival at the laboratory, an intravenous catheter was inserted into an antecubital vein, and a blood sample was collected (baseline). Then participants ingested one of the substances, and blood samples were collected every 25 min for a 100-min period. This time was chosen as it has been reported that both NH4Cl and NaHCO3 alter pH within 90 min of ingestion (35). Participants were asked to complete a validated questionnaire of gastrointestinal discomfort at baseline and 100 min after the ingestion (27).
As a high dose of NH4Cl (~0.3 g·kg−1 BM) might result in strong gastrointestinal discomfort (25,37), we tested the effects of a reduced total amount given in two small doses (0.08 + 0.07 g·kg−1 BM), as an attempt to promote significant alterations in blood acid–base balance without any gastrointestinal discomfort. Therefore, participants ingested 0.08 g·kg−1 BM of NH4Cl at baseline and then an additional 0.07 g·kg−1 dose 50 min later (time “50” min), summing to a total dose of 0.15 g·kg−1 BM.
For NaHCO3 and CaCO3 ingestion, given that no severe gastrointestinal distress has been reported with either 0.3 g·kg−1 BM of NaHCO3 or 0.3 g·kg−1 BM of CaCO3 (placebo) (32,39), these doses were ingested as a single dose at baseline. However, to match the number of capsules consumed across the conditions, and therefore to blind the experimental conditions, participants ingested “neutral” capsules with microcrystalline cellulose (inert substance) at minute 50 in the NaHCO3 and placebo conditions. All substances were offered in gelatin capsules of the same size, shape, and color.
A 600-μL blood sample was collected in a specific syringe for blood gas analysis (BD A-line™ syringe with slip tip). The pH, HCO3− concentration ([HCO3−]), and base excess (BE) were determined by an automatic blood gas analyzer (RAPIDPoint® 350, Siemens, Germany).
Questionnaire of Gastrointestinal Discomfort
A validated questionnaire to identify potential gastrointestinal discomfort was administered (27). This questionnaire consisted of 15 items describing possible side effects related to the ingested substances, with values for each item ranging from one to 10, in which 1 correspond to “no problem at all” and 10 to “the worst it has ever been.” The symptoms were considered severe when the score was equal to or greater than 5 (27).
Data are expressed as mean ± SD, unless otherwise stated. A two-way repeated-measures ANOVA followed by the least significant difference test were used to verify the effects of condition and time on pH, [HCO3−], and BE. Analyzes were conducted using SPSS (13.0) software, and the statistical significance was accepted at P ≤ 0.05.
Fifteen recreationally trained cyclists were invited to participate in this study. Four cyclists did not fully complete all tests for personal reasons. Thus, eleven cyclists (age = 35.7 ± 7.1 yr, mass = 74.7 ± 10.0 kg, height = 1.75 ± 0.10 m, peak oxygen uptake [V˙O2peak] = 3.99 ± 0.48 L·min−1 [54.1 ± 9.3 mL·kg−1·min−1], peak PO [PPO] = 304 ± 30 W) completed the study. Participants trained (~135 ± 47 min·d−1) and competed regularly in regional competitions and were all accustomed to TT exercise. Some of them participated in previous studies of our laboratory involving 4-km cycling TT. They were classified as recreationally trained cyclists in accordance with their V˙O2peak and PPO (13). All study procedures, as well as protocol, benefits, and risks, were explained before the commencement of trials, and participants gave their written informed consent. This study was approved by the Ethics Committee of the University of São Paulo.
Participants attended the laboratory on six different occasions. During the first visit, participants answered a questionnaire on readiness for physical activity readiness questionnaire and underwent a medical examination and an electrocardiogram. Only after medical clearance did participants begin the exercise tests. Following the measurement of mass and height, participants performed an incremental test to establish their V˙O2peak and PPO. During the second and the third visits, participants were familiarized with the 4-km TT and experimental procedures. On subsequent visits (visits 4 to 6), participants performed a 4-km TT 100 min after ingesting NH4Cl, NaHCO3, or CaCO3 using a double-blind, counterbalanced, and repeated-measures design. A 3-d minimum interval and a 7-d maximum interval were used between the experimental sessions to wash out any residual effect of fatigue and/or substance ingestion. Participants were asked to refrain from vigorous physical activity, alcohol, tobacco, and caffeine for 24 h before each experimental session. During the 24 h before the first familiarization session, the participants recorded all foods consumed (type, amount, and time), and then this record was photocopied and returned to them; this food intake was replicated during the 24 h before each subsequent trial. Participants consumed their last meal 2 h before all familiarization and experimental sessions. All sessions were conducted at the same time of day to avoid potential circadian variation on pacing strategy (16). Performance times and treatment order were not revealed to the participants until the last trial had been completed.
All tests were performed on the participant's own bicycle attached to a CompuTrainer (RacerMate®, Seattle, WA), which was calibrated before each test in accordance with the manufacturer's recommendations. The V˙O2, carbon dioxide production (V˙CO2), ventilation (V˙E), tidal volume (VT), respiratory frequency (fR), end-tidal oxygen partial pressure (PetO2), end-tidal carbon dioxide partial pressure (PetCO2), V˙E/V˙O2, and V˙E/V˙CO2 were measured breath-by-breath during the trials via a computerized system (Cortex Metalyzer 3B; Cortex Biophysik, Leipzig, Germany). The participants wore a face mask of appropriate size and breathed through a low dead space of 70-mL low-resistance piece connected to a bidirectional digital volume transducer turbine (range = 0.05–20 L·s−1, resolution = 7 mL, accuracy = ±2%). The inspired and expired gas volume and gas concentration signals were continuously sampled through a capillary line connected to the mouthpiece, with the gas concentration determined using electrochemical cell (O2, range = 0–100 vol%, accuracy = ±0.1 vol%) and ND infrared (CO2, range = 0–13 vol%, accuracy = ±0.1 vol%) analyzers (Metalyzer 3B, Cortex Biophysik). The gas analyzer was calibrated according to the manufacturer's specifications before each test (Cortex Metalyzer 3B instruction manual) using room air and a certified standard gas mixture (12% O2 and 5% CO2; White Martins Praxair Inc., Sao Paolo, Brazil). The turbine for inspired volume was calibrated with a 3-L syringe (Hans Rudolph, Kansas City, MO).
After a 3-min warm-up at 100 W, 30-W increments were applied every 3 min in a square-wave manner until the participant could no longer maintain the required pedal frequency (80–90 rpm). V˙O2peak was defined as the average V˙O2 during the last 30 s of the test. PPO was determined as the highest PO achieved during the last fully completed stage. When the participants could not maintain the PO during the entire last stage, the PPO was calculated using the fractional time completed in the last stage multiplied by the increment rate. Participants were instructed to record their RPE during each stage to familiarize with the Borg 15-point category scale.
In the first familiarization trial, participants performed three 10-s simulated starts to fix the initial gear ratio. Once fixed, they replicated this initial gear ratio during the subsequent trials. The participants were free to adjust the gear and pedal frequency once the TT had started (after ~2 s). They performed the familiarization sessions simulating the 4-km TT with the same procedures used in the experimental sessions. The RPE was recorded each 1 km so that participants were fully familiarized with the Borg 15-point category scale. The times to complete the 4-km TT in the two familiarization sessions were not significantly different (familiarization 1 = 383.0 ± 19.5 s, familiarization 2 = 384.7 ± 20.6 s, typical error = 1.40%). A 4-km cycling TT was chosen because there is evidence that both NH4Cl and NaHCO3 affect performance during exercise of this duration (21,28,35).
The same procedures as the preliminary study were adopted for substance ingestion. However, taking into account the results of the preliminary study, participants ingested 0.15 g·kg−1 BM rather than 0.3 g·kg−1 BM of CaCO3 in the placebo condition (see the results of the preliminary study for more details).
Experimental tests were conducted at the same time of day in a stable laboratory environment (23.8°C ± 2.9°C and 56.6% ± 3.6% relative humidity). After a catheter was inserted into a forearm vein, participants remained in a quiet sitting position for at least 30 min. They then ingested one of the substances and remained sitting at rest for a further 100 min. Thereafter, they performed a 5-min warm-up at 150 W followed by a 2-min break. Participants were then asked to complete the 4-km TT as fast as possible. The PO was recorded at a frequency of 1 Hz. Participants wore a mask throughout all experimental trials to allow analysis of respiratory parameters. Feedback about elapsed distance was given every 200 m, but participants were not informed about velocity, cadence, or heart rate. The participants were asked to report an overall RPE every 1 km using cues derived from all sensations experienced during exercise, with the verbal cues ranging from “6 = no exertion at all” to “20 = maximal exertion” (6). Participants received verbal instructions previous each trial (6) and were asked with a standardized question “How hard do you feel this exercise is?” The scale was visible to the participants during the whole trials. Participants were not verbally encouraged throughout the trials to ensure consistent conditions between trials.
Blood Collection and Analysis
Venous blood sample (600 μL) was obtained at baseline and 100 min after substance ingestion (i.e., pre-TT), and every 1 km during the TT for blood pH, [HCO3−], and BE measurements. Additional 500-μL venous blood samples were collected at the same time points, and then 25 μL was drawn off and transferred into microtubes containing 25 μL of sodium fluoride (1%) and centrifuged at 3000 rpm (5°C) for 5 min. Plasma lactate concentration [La] was measured in duplicate using an enzymatic colorimetric reaction read in a spectrophotometer (546-nm wavelength, EON; BioTek Instruments Inc., Winooski, VT). Data from our laboratory showed a within-subject coefficient of variation around 1.35% for blood pH, [HCO3−], and BE measurements. The corresponding CV for [La] was 1.96% in the present study.
The aerobic (Paer) and the anaerobic (Pan) mechanical PO (reflecting the mechanical aerobic and anaerobic energy expenditure rates, respectively) were estimated from the metabolic power and gross mechanical efficiency, as previously described in detail (20). Briefly, the metabolic power was first calculated during the warm-up from the following equation:
where V˙O2 is the oxygen uptake in liters per minute.
The gross mechanical efficiency was determined by dividing PO at warm-up (i.e., 150 W) by metabolic power and then corrected for high-intensity exercise (12). The Paer during TT was calculated by multiplying the metabolic power estimated during the TT (from equation 1) by gross mechanical efficiency (20). We assumed an RER equal to 1.00 to calculate metabolic power during the TT (20). The Pan was calculated by subtracting the calculated Paer from the total measured mechanical PO (20). Total energy expenditure (EEtot) and total aerobic and anaerobic energy expenditure (EEaer and EEan, respectively) were computed by calculating the area under the PO, Paer, and Pan versus time curves, respectively (38).
The individual RPE values were regressed against distance (km) and the slope was computed using a least squares fitting procedure (Origin, Microcal, Piscataway, NJ).
Questionnaire of Gastrointestinal Discomfort
Participants reported any sense of gastrointestinal discomfort using the same questionnaire and procedures described in the preliminary study.
Because of the lack of data regarding the effects of NH4Cl and NaHCO3 on 4-km TT performance, the required sample size was estimated using the standardized effect size (ES = 0.5) reported in a meta-analysis of the effects of NH4Cl and NaHCO3 on exercise performance during constant-load exercise (7). With an alpha of 0.05 and a desired power of 0.95, the total effective sample size necessary to achieve statistical significance was estimated to be 12 participants. However, assuming that 20% would not fully comply with instructions or would drop out during the data collection, the starting sample size was increased to 15 participants. The power using our final sample size (n = 11) and the overall effect size of substances on mean PO (partial eta-squared ηp2 = 0.41) was 0.88. All of these calculations were performed using G-Power software (version 3.1.7) (15).
Data are expressed as mean ± SD, unless otherwise stated. Data distribution was verified by Shapiro–Wilk’s test. A two-way repeated-measures ANOVA followed by a least significant difference test was used to verify the effects of condition and distance on dependent variables (pH, [HCO3−], BE, [La], respiratory parameters, PO, Paer, and Pan). Means of performance time, PO, Paer, Pan, EEtot, EEaer, and EEan and RPE slope were compared between experimental conditions using a one-way repeated-measures ANOVA followed by the least significant difference test. Gastrointestinal discomfort was analyzed only qualitatively and reported descriptively. All analyzes were conducted using SPSS (13.0) software and the statistical significance was accepted at P ≤ 0.05.
Blood pH, [HCO3−], and BE
There were significant main effects of time, condition, and interaction for blood pH, [HCO3−], and BE (P < 0.05; Figs. 1A–C). In the NaHCO3 and NH4Cl conditions, the pH remained stable for the first 50 min (P > 0.05); thereafter, compared with baseline, the pH increased and decreased significantly (P < 0.05) until 100 min, respectively. No significant differences were found between time points for placebo (P > 0.05). At baseline and 25 min postingestion, the pH values were similar between conditions (P > 0.05). However, from minute 25 to minute 100, the pH values were significantly higher in NaHCO3 than that in placebo and NH4Cl (P < 0.05), whereas the pH was lower in NH4Cl than that in placebo (P < 0.05). Similarly, both [HCO3−] and BE increased in NaHCO3 and decreased in NH4Cl from minute 25 to minute 100 compared with baseline (P < 0.05). [HCO3−] and BE were higher at minutes 75 and 100 compared with baseline for placebo (P < 0.05). At baseline, [HCO3−] and BE were lower in the NaHCO3 condition than that in the NH4Cl condition (P < 0.05), but both were not different from placebo (P > 0.05). Values for [HCO3−] and BE from minute 50 to minute 100 were significantly higher in NaHCO3 compared with placebo and NH4Cl and significantly lower in NH4Cl compared with placebo (P < 0.05).
There were no participant reported values equal to or higher than five points (i.e., severe) for all 15 items of the gastrointestinal discomfort questionnaire at both baseline and 100 min after substance ingestion for any experimental condition, except for one participant, who vomited 35 min after NH4Cl ingestion (the session was then interrupted). One week later, we replicated the same protocol in this participant, and he was able to complete all experimental procedures, but some symptoms such as stomach problems, nausea, heartburn, and urge to vomit were reported at the end of 100 min (five points stated for all symptoms).
Because of the findings of this preliminary study, the CaCO3 dose was reduced in the main study to avoid any increase in [HCO3−] and BE over time.
Blood pH, [HCO3−], BE, and plasma [La]
Significant main effects of condition, time, and interaction effect (P < 0.05) were found for blood pH, [HCO3−], and BE (Figs. 2A–C). The pH, [HCO3−], and BE values were similar between conditions at baseline (P > 0.05), but 100 min after ingestion (pre-TT), they increased in NaHCO3 (P < 0.05), remained stable in placebo (P > 0.05), and decreased in NH4Cl (P < 0.05). During the trial, the pH, [HCO3−], and BE decreased progressively in all conditions compared with pre-TT (P < 0.05). However, the pH throughout the trial was significantly lower in NH4Cl compared with placebo and NaHCO3 (P < 0.05), except for the 3-km interval, in which NH4Cl was only lower than NaHCO3 (P < 0.05). The pH in NaHCO3 was significantly higher than placebo at pre-TT and during the first and last kilometers of the TT (P < 0.05). [HCO3−] and BE were significantly lower at pre-TT and 1 km for NH4Cl compared with placebo and NaHCO3 (P < 0.05) but were higher in the NaHCO3 compared with placebo (P < 0.05). However, from kilometers 2 to 4, [HCO3−] and BE were significantly lower in NH4Cl compared with NaHCO3 (P < 0.05), but without significant differences between placebo and the other two experimental conditions (P > 0.05).
There was a main effect of condition and time (P < 0.05) but no interaction effect for plasma [La] (P > 0.05; Fig. 2D). The [La] remained stable at baseline and pre-TT (P > 0.05), but then it increased during the trial in all three conditions (P < 0.05). Plasma [La] was systematically lower in NH4Cl than that in placebo and NaHCO3 throughout the TT (P < 0.05), but there were no differences between NaHCO3 and placebo conditions (P > 0.05).
There were significant main effects of time, condition, and interaction for V˙O2 and V˙CO2 (P < 0.05; Figs. 3A and B). V˙O2 and V˙CO2 increased progressively up to 2 km (P < 0.05) and then remained stable in all conditions (P > 0.05), but V˙O2 was lower in NH4Cl compared with NaHCO3 during the first 3 km, whereas V˙CO2 was lower than both placebo and NaHCO3 at kilometers 2 and 3 (P < 0.05). There was also a main effect of time and interaction (P < 0.05) but not condition for PetO2 (P > 0.05; Fig. 3C). The PetO2 increased progressively up to 2 km (P < 0.05) and then remained relatively stable (P > 0.05). The PetO2 was higher in NH4Cl compared with the other two conditions only at pre-TT (P < 0.05) and higher in NH4Cl than NaHCO3 only at kilometer 1 (P < 0.05). For PetCO2, there was a main effect of time, condition, and interaction (P < 0.05; Fig. 3D). The PetCO2 decreased progressively during the trial in all conditions (P < 0.05), but values in NH4Cl were lower than both placebo and NaHCO3 from pre-TT to kilometer 3 (P < 0.05) and lower than NaHCO3 at kilometer 4. Values of PetCO2 were also higher in NaHCO3 than that in placebo at pre-TT and kilometer 1 (P < 0.05).
There was a significant main effect of time (P < 0.05), but not of condition, or an interaction effect (P > 0.05) for V˙E, VT, fR, and V˙E/V˙O2 (Figs. 3E–H). These respiratory parameters increased up to kilometer 2 (P < 0.05) and then remained stable (P > 0.05). Main effects were found for time and condition (P < 0.05) for V˙E/V˙CO2, but there was no interaction effect (P > 0.05; Fig. 3I). The V˙E/V˙CO2 increased progressively throughout the trial in all conditions (P < 0.05) but was higher in NH4Cl compared with placebo and NaHCO3 (P < 0.05), with no differences between placebo and NaHCO3 (P > 0.05).
PO and energy system distribution
There were significant main condition and distance effects for PO, Pan, Paer, and RPE (P < 0.05) but no interaction effect (P > 0.05, Figs. 4A–D). The PO in NH4Cl was lower throughout the TT when compared with placebo and NaHCO3 (P < 0.05), but there were no differences between NaHCO3 and placebo conditions (P > 0.05). For all conditions, PO values in the first and last kilometers were higher compared with the second and the third kilometers (P < 0.05). The Pan mirrored the PO profile during the trial. The Paer increased progressively during the first 2 km (P < 0.05) then remained relatively stable in all experimental conditions (P < 0.05), but it was always lower in NH4Cl compared with placebo (P < 0.05), with no difference between NH4Cl and NaHCO3 or placebo and NaHCO3 (P > 0.05). The RPE also increased throughout TT in all conditions (P < 0.05), with higher values in NH4Cl compared with NaHCO3 (P < 0.05), but without differences between placebo and the other two conditions (P > 0.05). However, the rate of increase in RPE over the distance was similar between the conditions (placebo = 2.3 ± 1.0, NH4Cl = 2.3 ± 1.0, NaHCO3 = 2.5 ± 1.0 U·km−1, P > 0.05).
Overall performance and total amount of aerobic and anaerobic energy expenditure
Mean PO and performance time were impaired in NH4Cl when compared with both placebo and NaHCO3 (P < 0.05), with no difference between NaHCO3 and placebo (Table 1, P > 0.05). This was accompanied by a lower mean Pan, EEan, and EEtot in NH4Cl than both placebo and NaHCO3 (P < 0.05), with no difference between NaHCO3 and placebo (P > 0.05). Furthermore, the mean Paer was significantly lower in NH4Cl than that in placebo (P < 0.05), but there was no difference between NH4Cl and NaHCO3 or placebo and NaHCO3 (P > 0.05). The EEaer was similar across the conditions (P > 0.05).
No participant reported values equal to or higher than five points (i.e., severe symptom) for all 15 items of the gastrointestinal discomfort questionnaire at baseline for any experimental condition. In the NaHCO3 condition, one participant reported symptoms related to diarrhea (five points) at 100 min postingestion, but he was able to perform the trial without stopping. In the NH4Cl condition, one participant vomited around 15 min and another one around 60 min after the NH4Cl ingestion. For these two participants, the test was cancelled and repeated 1 wk later without any severe gastrointestinal discomfort (all items lower than five points). No participant reported any symptom above five points after CaCO3 ingestion in the placebo condition at any time point.
The main finding of this study was that induced acidosis causes a significant reduction in both the PO and the anaerobic energy expenditure rate during a 4-km cycling TT. On the other hand, induced metabolic alkalosis had no effect on PO or anaerobic energy expenditure rate. Our results suggest that PO during a 4-km cycling TT and, consequently, anaerobic metabolism, might be regulated by a reduction, but not an increase, in blood pH before exercise. Furthermore, we observed that a reduced and fractionated dose of NH4Cl is effective in reducing the blood pH without severe gastrointestinal distress, suggesting the observed effects were independent of gastrointestinal discomfort.
During the 4-km cycling TT, regardless of condition, the PO–distance curve revealed a classic U-shaped pacing strategy (11,38) (Fig. 4A). However, confirming our hypothesis that induced metabolic acidosis would impair the performance, participants exercised at a lower PO throughout the trial in NH4Cl compared with the placebo and NaHCO3 conditions (Fig. 4A). This reduction in PO during the trial with NH4Cl was accompanied by a large reduction in the Pan (~14%, Fig. 4B), which led to underutilization of the anaerobic reserve (~13% reduction in EEan, Table 1), and to a lesser extent to a lower Paer (~4%, Fig. 4C). Metabolic acidosis also was associated with lower exercise V˙O2, V˙CO2, and PetCO2 (Figs. 3A–D). Interestingly, V˙E, VT, and fR response (Figs. 3E–G) were similar between experimental conditions, and RPE pattern of increase was not different between NH4Cl and placebo (Fig. 4D). This suggests that inducing acidosis before the exercise promoted a reduction in exercise intensity and attenuated the physiological response during the trial.
Although our data cannot precisely provide the main mechanism by which acidosis reduced PO during the 4-km cycling TT, these results may be accommodated by a recent theoretical concept termed sensory tolerance limit (26). A reduced muscle/blood pH increases sensory feedback signals, accounting for an important part of the sum of all feedback within the sensory areas in the CNS (1). If participants are not able to change PO, as occurs during constant-load exercise, compensatory mechanisms such as a greater buffering of “extra” H+ by HCO3 would cause greater V˙CO2 and PetCO2, as predicted from the classic model of acid–base regulation (22). This would be associated with an increase in V˙E due to an increase in fR and/or VT. Inevitably, these would exacerbate afferent signals from active muscles and respiratory muscles, ultimately causing earlier fatigue and exercise interruption. However, during a TT, in which cyclists can adjust their PO (10), increased afferent signals from muscle/blood pH to sensory areas into the brain might reduce descending neural drive to the active muscles to ensure the trial can be covered before an early metabolic disturbance. In fact, in the present study, we found that even with a reduction in blood pH, [HCO3−], and BE after the induced metabolic acidosis (Figs. 2A–C), there was a decrease rather than an increase in the V˙CO2 and PetCO2, whereas the V˙E, VT, and fR response were not altered. This suggests that the “attenuated” metabolic response after NH4Cl ingestion, as demonstrated by lower V˙O2, V˙CO2, and PetCO2, was a consequence rather than a cause of reduced PO. This is also supported by an unchanged RPE response between acidosis and placebo conditions, which corroborate studies suggesting that under unfavorable conditions during a trial, such as reduced muscle glycogen (30), undesirable feelings cause participants to reduce their PO while maintaining the same RPE pattern during the trial.
Alternatively, the attenuated PO during the trial with NH4Cl could simply be explained by a peripheral, energy-limited mechanism. The large reduction in the Pan (~14%) led to a lower EEan (~13%). Considering the total anaerobic reserve (i.e., EEan) is assumed to be fixed and limited in capacity (9,19), our results could indicate that a reduction in pH might limit the use of the anaerobic capacity by a peripherally located mechanism. NH4Cl increases extracellular [H+], which reduces the H+ gradient across the sarcolemma, reducing H+/La− monocarboxylate transporter activity and the efflux of H+ from the muscle to the blood during exercise (8,23). A reduction in blood pH of close to 0.08 U after NH4Cl ingestion, similar to that found in the present study (~0.07), is sufficient to induce a lower intramuscular pH during exercise (8,14,23). This may inhibit muscle glycogenolytic and glycolytic flux during exercise because of the inhibition of glycogen phosphorylase and/or phosphofructokinase, respectively (23). Although we have no data in the present study to prove or disprove this possibility, it seems to be improbable that an exclusively peripherally controlled mechanism could account for entire reduction in PO once that experimental evidences point that reduction in H+ and associated alterations play an important role in activate group III/IV sensory neurons (5).
A challenge to explain pH regulating the PO during a 4-km cycling TT via afferent signals, or alternatively by a direct regulation of peripheral anaerobic metabolism, is that contrary to our second hypothesis, previous metabolic alkalosis did not increase PO and overall performance. Although this is consistent with the findings of previous studies reporting no improvements in exercise performance during a high-intensity cycling TT with NaHCO3 (24,39), this is contrary to another study showing an increase in performance during a 3-km cycling TT after NaHCO3 ingestion (29). The reason for this discrepancy is not clear but may be related to factors such as the distance/duration of the trial, physical fitness of participants, consumed dose, and the time between the beginning of ingestion and the beginning of the trial. Nevertheless, consistent with a nonimproved performance with NaHCO3 ingestion, there was no effect of alkalosis on the Pan (Fig. 4B), EEan (Table 1), Paer (Fig. 4C), and other metabolic and respiratory parameters, except an increase in PetCO2 (Fig. 3). These almost identical metabolic and physiological responses between NaHCO3 and placebo ingestion, even with an enhanced pH throughout the trial in NaHCO3, suggest that the sensory tolerance limit may not have been sufficiently attenuated. In addition, although a reduction in pH may increase afferent signaling, it is not apparent that an increase in pH above rest value should reduce it. It has been previously suggested that group III/IV afferent nerve fibers seems to respond the infusion of a metabolite combination mimicking intense exercise, including protons accumulation (33), but to best of our knowledge, it is unknown if these afferent fibers would decrease firing rate if protons were reduced. Another possibility is that muscle pH may not have been sufficiently increased during exercise with NaHCO3 ingestion. Previous studies have demonstrated that NaHCO3 ingestion up to blood pH reaches 7.55 U does not affect intracellular pH (4). The mean blood pH after NaHCO3 ingestion in the present study was 7.44 (ranging from 7.36 to 7.49). A higher dose of NaHCO3 could have caused a larger increase in muscle pH, but it should be taken in account that higher doses of NaHCO3 would promote severe gastrointestinal discomfort (31).
Although the RPE increased similarly between placebo and metabolic acidosis and alkalosis, RPE was lower in NaHCO3 than that in NH4Cl (Fig. 4D). Although there was no interaction between distance versus condition, a closer view in RPE response (Fig. 4D) reveals that the larger differences occurred mainly in the first kilometer, with differences becoming smaller for the following kilometers. It is difficult to explain why this slightly reduced RPE was not converted to a better performance, although performance was better in NaHCO3 compared with NH4Cl and similar between NaHCO3 and placebo. These results may suggest that either the participants were not familiarized enough with TT and therefore were unable to realize the PO should have increased or RPE differences was too small that participants could not advance for a higher PO in NaHCO3 condition. Although participants of our study were not highly trained cyclists and not specialized in 4-km cycling TT, they were accustomed to TT and part of them had performed several 4-km cycling TT in previous studies of our laboratory. In fact, we found a small typical error between two trials (<1.5%), which suggests that participants were familiar with this kind of task. However, a possible long-term (yr) effect of learning 4-km cycling TT cannot be fully disregarded. On the other hand, it seems to be more probable that the RPE differences between conditions were not large enough to be converted in a higher PO with NaHCO3. In fact, a similar RPE response between NaHCO3 and placebo was accompanied by a similar performance, whereas a slightly reduced RPE in NaHCO3 in comparison with NH4Cl was accompanied by a better performance in the former. Supporting this assumption, the rate of increase in RPE over the distance was similar between the conditions, suggesting that NaHCO3 did not affect the overall programming by which participants monitor the homeostatic disturbance against what is ultimately tolerable.
Methodological aspects of NH4Cl ingestion, such as dosage, administration form, and time course for desired acid–base balance changes, have not been completely standardized (25,36,37). As large (~0.30 g·kg−1 BM) doses of NH4Cl are associated with gastrointestinal complications (25,37), which could impair per se exercise performance, we first tested the efficacy of a reduced, split NH4Cl dose. This was successful at achieving significant reductions in blood pH, [HCO3−], and BE before and during the exercise, similar to those reported in previous studies using higher doses (25,37), but with less gastrointestinal discomfort (only in two participants in our main study, both of whom subsequently repeated the trial and reported no severe gastrointestinal discomfort). One participant reported gastrointestinal discomfort consuming the well-established NaHCO3 dose (29,37) but was still able to complete the trial successfully. Our modified CaCO3 dose (from 0.3 to 0.15 g·kg−1 BM, respectively) was also effective because no alterations from baseline to pre-TT for blood pH, [HCO3−], and BE were observed in our main study. Therefore, the NH4Cl, the NaHCO3, and the CaCO3 doses used in the present study successfully achieved blood acidosis, alkalosis, and neutrality, respectively, with minimal gastrointestinal discomfort. These methodological insights might be useful to inform further studies investigating the effects of acidosis and alkalosis on energy expenditure and exercise performance.
A possible limitation of the present study was the collection of blood samples and air samples during the trial, which may have affected the cyclist’s performance. However, to avoid any discomfort related to the catheter, we inserted the catheter in a vein located in the middle third of the forearm. This strategy did not restrict the movement of the upper limbs during the trial. We also added an extension tube to the catheter, which diminished the contact and possible interference of the researcher at the time of blood collection. Furthermore, the time to complete the 4-km TT in the placebo condition (384.9 ± 19.1 s) was very similar to that found in the two familiarization sessions (familiarization 1 = 383.0 ± 19.5 s, familiarization 2 = 384.7 ± 20.6 s), in which there was no blood collection or mask to collect respiratory gases. This suggests interference caused by the collection of blood or respiratory gases was unlikely. Although this also indicates that participants were well familiarized with TT, it is important to recognize that participants were recreationally trained cyclists but not specialized in 4-km cycling TT. Furthermore, electromyography activity was not monitored during the trial, which could provide important information about central motor drive during the trial. Further studies should test if this model could be reproducible in highly trained cyclists and if central motor drive is affected by acidosis and alkalosis. Finally, another potential limitation is that substance ingestion caused gastrointestinal discomfort in a few participants. However, no participant started the trial reporting gastrointestinal discomfort higher than five points on the gastrointestinal discomfort scale, a score considered a “threshold” for severe gastrointestinal discomfort that could impair the performance (27).
The results of the present study revealed that PO and anaerobic energy expenditure rate during a 4-km cycling TT may be regulated by a reduction but not increase in blood/muscle pH, which is independent of gastrointestinal discomfort. The induced preexercise metabolic acidosis seems to negatively effect on 4-km cycling TT performance by prevent full use of the anaerobic reserve.
Carlos Rafaell Correia-Oliveira is grateful to Coordination of Improvement of Personnel of Superior Level (CAPES) for his Ph.D. scholarship. The authors thank all of the cyclists who took part in this study. They are also grateful to Dr. Felipe Hardt for his excellent medical assistance during the experiments.
The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.
The results of the study are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation. No financial support was received. The authors declare no conflict of interest. The results of this study do not constitute endorsement by the American College of Sports Medicine.
1. Amann M, Dempsey JA. Locomotor muscle fatigue modifies central motor drive in healthy humans and imposes a limitation to exercise performance
. J Physiol
2. Amann M, Proctor LT, Sebranek JJ, Pegelow DF, Dempsey JA. Opioid-mediated muscle afferents inhibit central motor drive and limit peripheral muscle fatigue development in humans. J Physiol
3. Bishop D. Dietary supplements and team-sport performance
. Sports Med
4. Bishop D, Edge J, Davis C, Goodman C. Induced metabolic alkalosis affects muscle metabolism and repeated-sprint ability. Med Sci Sports Exerc
5. Blain GM, Mangum TS, Sidhu SK, et al. Group III/IV muscle afferents limit the intramuscular metabolic perturbation during whole body exercise in humans. J Physiol
6. Borg E, Kaijser L. A comparison between three rating scales for perceived exertion and two different work tests. Scand J Med Sci Sports
7. Carr AJ, Hopkins WG, Gore CJ. Effects of acute alkalosis and acidosis on performance
: a meta-analysis. Sports Med
8. Churchward-Venne TA, Kowalchuk JM, Marsh GD. Effects of ammonium chloride
ingestion on phosphocreatine metabolism during moderate- and heavy-intensity plantar-flexion exercise. Eur J Appl Physiol
9. Corbett J, Barwood MJ, Parkhouse K. Effect of task familiarisation on distribution of energy during a 2000 m cycling time trial. Br J Sports Med
10. Correia-Oliveira CR, Bertuzzi R, Dal'Molin Kiss MA, Lima-Silva AE. Strategies of dietary carbohydrate manipulation and their effects on performance
in cycling time trials. Sports Med
11. Correia-Oliveira CR, Santos RA, Silva-Cavalcante MD, et al. Prior low- or high-intensity exercise alters pacing strategy, energy system contribution and performance
during a 4-km cycling time trial. PLoS One
12. de Koning JJ, Noordhof DA, Uitslag TP, Galiart RE, Dodge C, Foster C. An approach to estimating gross efficiency during high-intensity exercise. Int J Sports Physiol Perform
13. De Pauw K, Roelands B, Cheung SS, de Geus B, Rietjens G, Meeusen R. Guidelines to classify subject groups in sport-science research. Int J Sports Physiol Perform
14. Edge J, Mündel T, Pilegaard H, et al. Ammonium chloride
ingestion attenuates exercise-induced mRNA levels in human muscle. PLoS One
15. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods
16. Fernandes AL, Lopes-Silva JP, Bertuzzi R, et al. Effect of time of day on performance
, hormonal and metabolic response during a 1000-M cycling time trial. PLoS One
17. Foster C, de Koning JJ, Hettinga F, et al. Pattern of energy expenditure during simulated competition. Med Sci Sports Exerc
18. Foster C, de Koning JJ, Hettinga F, et al. Effect of competitive distance on energy expenditure during simulated competition. Int J Sports Med
19. Hettinga FJ, de Koning JJ, Broersen FT, Van Geffen P, Foster C. Pacing strategy and the occurrence of fatigue in 4000-m cycling time trials. Med Sci Sports Exerc
20. Hettinga FJ, de Koning JJ, Meijer E, Teunissen L, Foster C. Biodynamics. Effect of pacing strategy on energy expenditure during a 1500-m cycling time trial. Med Sci Sports Exerc
21. Higgins MF, James RS, Price MJ. The effects of sodium bicarbonate
) ingestion on high intensity cycling capacity. J Sports Sci
22. Hirakoba K, Yunoki T. Blood lactate changes during isocapnic buffering in sprinters and long distance runners. J Physiol Anthropol Appl Human Sci
23. Hollidge-Horvat MG, Parolin ML, Wong D, Jones NL, Heigenhauser GJ. Effect of induced metabolic acidosis on human skeletal muscle metabolism during exercise. Am J Physiol
. 1999;77(4 Pt 1):E647–58.
24. Horswill CA, Costill DL, Fink WJ, et al. Influence of sodium bicarbonate
on sprint performance
: relationship to dosage. Med Sci Sports Exerc
25. Housh TJ, deVries HA, Johnson GO, Evans SA, McDowell S. The effect of ammonium chloride
and sodium bicarbonate
ingestion on the physical working capacity at the fatigue threshold. Eur J Appl Physiol Occup Physiol
26. Hureau TJ, Romer LM, Amann M. The ‘sensory tolerance limit’: a hypothetical construct determining exercise performance
? Eur J Sport Sci
. 2016; doi:10.1080/17461391.2016.1252428.
27. Jeukendrup AE, Vet-Joop K, Sturk A, et al. Relationship between gastro-intestinal complaints and endotoxaemia, cytokine release and the acute-phase reaction during and after a long-distance triathlon in highly trained men. Clin Sci (Lond)
28. Jones NL, Sutton JR, Taylor R, et al. Effect of pH on cardiorespiratory and metabolic responses to exercise. J Appl Physiol Respir Environ Exerc Physiol
29. Kilding AE, Overton C, Gleave J. Effects of caffeine, sodium bicarbonate
, and their combined ingestion on high-intensity cycling performance
. Int J Sport Nutr Exerc Metab
30. Lima-Silva AE, Correia-Oliveira CR, Tenorio L, Melo AA, Bertuzzi R, Bishop D. Prior exercise reduces fast-start duration and end-spurt magnitude during cycling time-trial. Int J Sports Med
31. McNaughton LR. Bicarbonate ingestion: effects of dosage on 60 s cycle ergometry. J Sports Sci
32. Noakes TD. Fatigue is a brain-derived emotion that regulates the exercise behavior to ensure the protection of whole body homeostasis. Front Physiol
33. Pollak KA, Swenson JD, Vanhaitsma TA, et al. Exogenously applied muscle metabolites synergistically evoke sensations of muscle fatigue and pain in human subjects. Exp Physiol
34. Poole DC, Burnley M, Vanhatalo A, et al. Critical power: an important fatigue threshold in exercise physiology. Med Sci Sports Exerc
35. Requena B, Zabala M, Padial P, Feriche B. Sodium bicarbonate
and sodium citrate: ergogenic aids? J Strength Cond Res
36. Robergs R, Hutchinson K, Hendee S, Madden S, Siegler J. Influence of pre-exercise acidosis and alkalosis on the kinetics of acid–base recovery following intense exercise. Int J Sport Nutr Exerc Metab
37. Siegler JC, Marshall P, Pouslen MK, Nielsen NP, Kennedy D, Green S. The effect of pH on fatigue during submaximal isometric contractions of the human calf muscle. Eur J Appl Physiol
38. Silva-Cavalcante MD, Correia-Oliveira CR, Santos RA, et al. Caffeine increases anaerobic work and restores cycling performance
following a protocol designed to lower endogenous carbohydrate availability. PLoS One
39. Vanhatalo A, McNaughton LR, Siegler J, Jones AM. Effect of induced alkalosis on the power-duration relationship of “all-out” exercise. Med Sci Sports Exerc