Effect of sodium bicarbonate contribution on energy metabolism during exercise: a systematic review and meta-analysis

The present article is a meta-analysis focusing on the contribution of sodium bicarbonate to energy metabolism during different types of exercise (i.e., aerobic-based and anaerobic-based). This study followed the Preferred Reporting Elements for Systematic Reviews and Meta-analysis (PRISMA) guidelines [25] and the eligibility criteria of articles was determined with the application of the Participants, Intervention, Comparison, Outcome and Study design (PICOS) question model [26], elements were used in title, abstract and/or full text of articles to identity studies that met the eligibility criteria (Table 4).

A systematic search was conducted using PubMed, Web of Science, SCOPUS, Medline and SPORTDiscus databases to identify eligible studies published from 2010 to June 2020. Search terms related with main concepts were used: “sodium bicarbonate” AND (“metabolism” OR “energy expenditure”) AND (“exercise” OR “physical activity” OR “sport”) AND “aerobic” AND “anaerobic”. Through this search, a total of 351 articles were obtained and 17 articles were finally included in this meta-analysis.

After obtaining the 351 articles according to the inclusion criteria of PICOS in the search, the following exclusion criteria were taken into consideration to determine the final studies: 1) Review and meta-analysis; 2) No sodium bicarbonate supplement ingestion or the outcomes measure not related to energy metabolism; 3) Supplement mixed with other supplements (i.e. caffeine or beta-alanine); 4) Animal experiments; 5) Injury participants or without training experience; 6) Study design not matched: not under the same experimental conditions (i.e., Hypoxia or ingested after high intensity exercise), without exercise after ingesting, no placebo as a comparison group; 7) Inadequate parameter measurement; 8) Data not described in detail (e.g., no mean or standard deviation (SD), no response after emailing author). The data collection process is presented in Fig. 1.

The methodological quality of the articles, was evaluated using McMaster’s Critical Review Form [27]. The McMaster Form contains 15 items that are scored depending on the degree to which the specific criteria were met (yes = 1, no = 0). A summary score was calculated for each article by summing the total score obtained across relevant items and dividing it by the total possible score. The evaluation score of the quality of the articles is shown in Table 5. The main deficiencies found in methodological quality are associated with item 14 of the questionnaire, which is “were drop-outs reported?”, as there is no description about whether participants dropped out or not.

Physiological results data were extracted in the form of mean, SD, and sample size for placebo and NaHCO₃ cohorts. Data were collected directly from tables or within text of the selected studies when possible. Data of 6 studies were partially abstracted by online graph digitizing software (WebPloDigitizer¹) when values were not reported in the text. This included values abstracted directly with mean and SD [28‐30] or calculated after obtaining mean and standard error (SE) [31, 32] or a 95% confidence interval (95% CI) [33]. A study was excluded from the meta-analysis when the missing data could not be provided, or the author did not respond [34‐36]. Dependent variables include those parameters relevant to energy metabolism after exercises following the supplement intervention. When pertinent data were not available or referenced in the article, the study was excluded from the meta-analysis.

The meta-analysis was conducted using the Review Manager 5.3 (v5.3, Cochrane Collaboration, Copenhagen, Denmark, 2020) in order to aggregate, via a random-effects model [37], the standardized mean difference (SMD) between the effects of NaHCO₃ and placebo cohorts. The mean ± SD and sample size were used to calculate SMD. A sub-group analysis was also performed to evaluate the influence on exercise with different metabolic characteristics. The use of the SMD summary statistic allowed all effect sizes to be transformed into a uniform scale, which was interpreted, according to Cohen’s conventional criteria [38], with SMD of < 0.20 being classified as negligible, 0.20–0.49 classified as small; 0.50–0.79 classified as moderate; and > 0.80 classified as large. Heterogeneity was determined using I² value, with values of 25, 50 and 75 indicating low, moderate and high heterogeneity, respectively. The results are reported as weighted means and 95% CI. The statistical significance was set at p < 0.05.

A total of 351 articles were initially identified through databases. Of the 186 that remained after the removal of 165 duplicates, 101 articles were not considered relevant and were excluded. Based on the inclusion criteria, 17 articles, published between 2010 and 2019, met the full set of criteria and were included for review. All descriptions and characteristics of the review studies are presented in Tables 6 and 7. Moreover, the quality assessment of selected articles was classified as Very Good (Table 5).

The study design, testing parameters and participants’ characteristics for the meta-analyzed studies are displayed in Tables 6 and 7. All studies are divided into two types of exercise, either anaerobic-based or aerobic-based. Exercise characteristics depend on the experimental design after NaHCO₃ intervention in these studies, which is whether the exercise is dominated by anaerobic or aerobic ability. After review, 11 articles [28, 29, 31‐33, 39‐44] were found to belong to anaerobic-based exercise for analysis of AnMS, which are the ATP-CP and glycolytic systems), and 6 articles [30, 45‐49] were found to belong to aerobic-based exercise for analysis of the oxidative system.

The total number of participants across all studies was 215. Studies either used mixed-sex samples (3 studies) or included only men (10 studies) or only women (1 study) and another 3 studies did not describe the gender of sample subjects. Out of the 17 included studies, 14 used a NaHCO₃ dose of 0.3 g•kg^− 1, two studies used the dose of 0.5 g•kg^− 1, and one study used the dose of 4 mmol•kg^− 1 (about 0.336 g•kg^− 1). The timing of ingestion ranged from 60 min up to 4 h pre-exercise. In some studies, the dose of NaHCO₃ was provided at one timepoint, with other studies splitting up the total dose at multiple timepoints. The duration of NaHCO₃ administration was either once or on 5 consecutive days. The type of administration was via gelatin capsules or tablets, but some studies did not report this information (Tables 6 and 7).

Metabolic by-products (e.g., lactic acid) are largely accumulated following the AnMS energy generation process. In the process of dissociating the metabolic by-product, the concentration of H⁺ in body fluids will increase and therefore lower the pH value. In order to reduce the effect of free H⁺, the alkaline substances in blood and muscle will combine with H⁺ to buffer or neutralize H⁺.

Fortunately, cells and body fluids have buffers such as HCO₃⁻, that can reduce the impact of H⁺. Without the buffers, H⁺ would lower the body’s pH value by 1.5, resulting in cell death. When the intracellular pH value is lower than 6.9, it inhibits the activity of important glycolytic enzymes and reduces the rate of glycolytic and ATP production. When the pH value reaches 6.4, H⁺ will stop any further decomposition of glycogen, causing ATP to rapidly decline until the end of the failure. However, due to the body’s buffering capacity, even during the most strenuous exercise, the concentration of H⁺ can be maintained at a very low level. Even when exhausted, the muscle pH value drops slightly from the steady state of pH 7.1, but it will not drop to a pH below 6.6–6.4 [1].

To sum up, ingesting NaHCO₃ will neutralize H⁺, thus affecting the content of buffer substances (HCO₃⁻) in the body and pH, thereby affecting the body’s acid-base balance. Since ingestion of NaHCO₃ leads to a higher efflux of lactate from the working skeletal muscle to the plasma, BLa can reflect metabolic ability to a certain extent. Therefore, the four variables (i.e., HCO₃⁻, pH, BE and BLa), at the last time point (i.e., the influence after the last exercise if it has two or more bouts, as with the variable used to analyze the oxidative system) were chosen to assess the influence of NaHCO₃ on AnMS.

When performing long-term moderate-intensity exercise, the ventilation volume matches the energy metabolism rate, and it is necessary to constantly change the ratio between the body’s oxygen uptake (VO₂) and carbon dioxide production (VCO₂). It is widely acknowledged that a higher VO₂ is associated with a stronger aerobic capacity. Most of the CO₂ (about 60–70%) produced during muscle exercise is transported back to the heart in the form of HCO₃⁻ [1]. CO₂ and water molecules combine to form carbonic acid, which is unstable and will soon dissolve, forming free H⁺ and HCO₃⁻:

When the blood enters the area where the partial pressure of carbon dioxide (PCO₂) in the lungs is low, H⁺ will combine with HCO₃⁻ to form carbonic acid, and then decompose into CO₂ and water:

After CO₂ enters the lungs, it is eliminated by dissociation, which is the main way to reduce H⁺ concentration when CO₂ is eliminated [1].

The amount and rate of gas exchange across the respiratory membrane are mainly determined by the partial pressure of each gas. The gas diffuses along the pressure gradient, from the part with the higher pressure to the lower pressure part. At standard atmospheric pressure, the partial pressure of oxygen (PO₂) outside the body is greater than that inside the body after alveolar gas exchange. When the exercising muscles require more oxygen to meet metabolic needs, the venous oxygen is depleted and accelerates the alveolar gas exchange, resulting in PO₂ reduction [1]. Therefore, O₂ enters the blood and CO₂ leaves the blood. PCO₂ is mainly used to determine whether there is respiratory acidosis or alkalosis. Increased PCO₂ suggests that there is insufficient lung ventilation, and CO₂ retention in the body, which leads to respiratory acidosis. Lower PCO₂, indicating hyperventilation (such as deeper or faster breathing), and excessive CO₂ elimination in the body, leads to respiratory alkalosis [1]. Therefore, an increase in PCO₂ will cause an increase in CO₂ in the blood, which will result in a decrease in the pH value.

To sum up, the change of O₂ and CO₂ during long-term moderate-intensity exercise can reflect aerobic capacity to a certain extent. For that reason, the four variables (i.e., VO₂, VCO₂, PO₂ and PCO₂) were chosen to assess the influence of NaHCO₃ on the oxidative system.

Results in the present analysis indicate that NaHCO₃ ingestion is effective in improving AnMS, which may be able to improve sport performance based on anaerobic capacity. The performance results of included studies showed that performance improved or was maintained the same when ingesting NaHCO₃, while a placebo showed a decline in sport performance. (Table 6). This result is different from that of other meta-analyses [52, 53], but similar to several individual studies which did not meet the present eligibility criteria [54, 55]. Two included studies [32, 33] reported no improvement in sport performance, and we found that the experimental exercise in these two articles were more likely based on the ATP-CP system to obtain energy (Table 6). This is similar to previous studies [12, 56], where the ATP-CP system was not affected by NaHCO₃ ingestion.

The key point of the contraindications in different results may be the gastrointestinal (GI) problems caused by ingestion of NaHCO₃. Because the bicarbonate buffer system is not solely responsible for blood pH and is also vital in other systems, such as the stomach and duodenum by neutralizing gastric acid, abdominal pain and diarrhea are often experienced by individuals who take NaHCO₃ [36, 57]. An article included in the present study also illustrated this problem [41]. While the results among all subjects indicated that the intake of NaHCO₃ has no effect on sports performance, after excluding subjects who had GI problems with ingestion of NaHCO₃, a significant difference in sports performance was observed. However, in this meta-analysis, the author extracted the data of all subjects from this article and verified that it did not affect the results of the meta-analysis. In response to the GI problems, some countermeasures have been taken that have been scientifically proven to alleviate or prevent GI problems. For example, ingestion of a large amount of water [58], with food [9], with carbohydrate [59] or administration as enteric-formulated capsules [60]. More measures to prevent GI problems may help demonstrate the improvement in sport performance with the intake of NaHCO₃ as subjects are not troubled by GI problems.