Acid-Base Balance in Healthy Adults: Beneficial Effects of Bicarbonate and Sodium-Rich Mineral Water in a Randomized Controlled Trial: The BicarboWater Study

Background Noncommunicable diseases (NCDs) are a global health challenge. The complex etiology of NCDs involves genetic, environmental, and lifestyle factors, including dietary habits. Chronic latent metabolic acidosis has been associated with an increased risk of NCDs. Alkalizing diets and mineral water consumption have shown promise in improving acid-base balance and potentially impacting NCDs. Methods In this randomized controlled intervention study, the effect of drinking 1,500–2,000 mL of mineral water daily on acid-base balance was evaluated. Ninety-four healthy participants were divided into two groups: one consumed mineral water with a high bicarbonate and sodium content (HBS, n = 49) and the other consumed mineral water with a low bicarbonate and sodium content (LBS, n = 45). Changes in venous blood gas and urinary acid-base parameters were measured over a short-term (3 days) and long-term (28 days) intervention period. Potential renal acid load (PRAL) and nutrient intake were calculated at baseline and after 28 days. Results HBS water consumption led to increased urinary pH (24-hour urine and spontaneous urine, both p < 0.001) and bicarbonate levels (p < 0.001), accompanied by reduced titratable acids (p < 0.001) and ammonium (p < 0.001), resulting in a lower renal net acid excretion (p < 0.001). These changes occurred in the short term and persisted until the end of the study. LBS consumption showed no significant effects on urinary pH but led to a slight decrease in bicarbonate (p < 0.001) and NH4+ (p < 0.001), resulting in a slight decrease in NAE (p=0.011). Blood gas changes were modest in both groups. Mineral water consumption in the HBS group altered dietary intake of sodium and chloride, contributing to changes in PRAL values. Conclusion The study demonstrates that the consumption of mineral water high in bicarbonate and sodium (1,500 mL–2,000 mL/day) can positively influence urinary acid-base parameters and reduce NAE, suggesting potential benefits in maintaining acid-base balance without adverse effects on human health. These findings highlight the importance of mineral water composition in acid-base regulation. This trial is registered with DRKS00025341.


Introduction
Noncommunicable diseases (NCDs) are a major global health challenge, afecting millions of people worldwide.Tey are among the leading causes of death and disabilityadjusted life years (DALYs) in the world [1,2].Te etiology of these diseases is complex and multifaceted, involving an interplay of genetic predisposition, environmental exposures, and lifestyle factors, particularly dietary habits.Several studies have shown a signifcant association between chronic latent metabolic acidosis and an increased risk of developing or progressing NCDs such as osteoporosis, type 2 diabetes, and even mental disorders or cancer [3][4][5][6][7].Prolonged adherence to an acid-forming diet results in a sustained disturbance of acid-base balance, leading to chronic latent acidosis.Terefore, maintaining a balanced acid-base status through dietary and lifestyle modifcations may have a positive impact on health [4,8].
In this context, dietary patterns rich in alkalizing components have demonstrated their ability to exert a favorable infuence on human acid-base balance [9][10][11][12].Furthermore, the consumption of mineral water has emerged as a promising way to improve acid-base balance [10,13].Te bioactive components present in mineral water, which include essential minerals such as calcium, magnesium, potassium, and sodium, are known to improve the potential renal acid load (PRAL) and counteract the acidogenic components in the diet [10,14].Some mineral waters also contain high levels of bicarbonate, which is also a natural component of the body's bicarbonate bufering system and plays a key role in neutralizing acids and maintaining the body's acid-base balance [15,16].Consistent with these fndings, consumption of bicarbonate-rich mineral water and a low PRAL value have been shown to exhibit benefcial efects on markers of acid-base metabolism [10].
Nevertheless, the existing body of evidence on the efect of mineral water on urinary and blood parameters related to acid-base status remains limited.Previous studies have focused on mineral waters with diferent bicarbonate concentrations and PRAL values or just alkalizing water.Tey also mainly investigated the efects of mineral water consumption over longer time periods (2-4 weeks).Te present intervention study was carried out with the aim of determining the infuence of two diferent mineral waters on acid-base balance in healthy, omnivorous adults, with comprehensive evaluations conducted over both short-term (3 days) and long-term (4 weeks) intervention periods.Tese waters were characterized by either a very high bicarbonate content or a very low PRAL value (HBS) or a relatively low bicarbonate content and a moderate PRAL value (LBS).

Study Design.
Tis parallel-group randomized controlled intervention study was designed to compare the infuence of mineral water very high in bicarbonate and high in sodium (HBS) with mineral water low in bicarbonate and sodium (LBS) on urinary and blood parameters of the acidbase-status.Te study was carried out at the Institute of Food Science and Human Nutrition at Leibniz University Hanover in Germany.It was conducted between June 2021 and May 2022.

Subjects.
Ninety-four healthy subjects, ranging in age from 30 to 65 years, were recruited from the general population of Hanover and Hildesheim.Primary inclusion factors were an omnivorous diet and a BMI within the range of 20.0 to 29.9 kg/m 2 .Subjects with chronic conditions such as manifest cardiovascular or renal diseases, as well as those under medical treatment for hypertension, were not considered for participation.

Ethical Approval.
Te study was conducted in accordance with the ethical standards described in the Declaration of Helsinki.Te study protocol was approved by the Ethics Committee of the Medical Chamber of Lower Saxony (Hanover, Germany) (BO/24/2021).After being fully informed of the purpose of the study and any potential adverse efects, all participants gave informed consent.Te study was registered in the German Clinical Trial Register (DRKS00025341, https://drks.de/search/en).In adherence to established protocols for the documentation of clinical trials, the CONSORT (Consolidated Standards of Reporting Trials) checklist was implemented to ensure transparency and comprehensiveness in explaining both the methodology and results of the trial (Supplementary S1).

Procedure and Test Products.
Te study consisted of a screening and a 4-week intervention phase.Te intervention included an initial examination at baseline (t 0 ) and two further examinations after 3 days (t 3 ) and after 4 weeks of mineral water consumption (t 28 ).Short-term efects were investigated at t 3 , while long-term efects were determined at t 28 .Participants received the bottles of the test water for the entire study period on the frst day of the intervention (t 0 ), immediately after the examinations.Alternatively, the water was delivered to their homes in advance.Participants who received the water in advance were instructed not to start drinking it until the end of the examinations on the frst examination day (t 0 ).
Participants were randomly assigned by an independent researcher to one of the two intervention groups using stratifed randomization according to the covariates sex and age (in descending order).Subjects assigned to the high bicarbonate and sodium group (HBS, n � 49) consumed a mineral water with a very high bicarbonate content (4,368 mg/L) and a high sodium content (1,708 mg/L).Conversely, those assigned to the low bicarbonate and low sodium group (LBS, n � 45) consumed a mineral water with a low bicarbonate (228 mg/L) and a low sodium (8.4 mg/L) content.Te composition of the test products is shown in Table 1.Over the course of the 4-week intervention, participants were instructed to consume a minimum of 1,500 mL and a maximum of 2,000 mL of the provided test water daily.Additional fuencies as tap water in its pure form, tea, cofee, mixture of juice and tap water were allowed.Besides that, subjects were advised to maintain their usual diet and maintain their physical activity level.To monitor the consumption of the test water, participants were asked to complete a daily protocol documenting the amount of water consumed daily.Tis protocol was used to check compliance with the study protocol and to calculate the amount of daily test water consumption after the intervention period.Moreover, compliance was checked by a question in the questionnaire at the end of the study.Participants were considered compliant if they drank 1,500 to 2,000 mL of the respective mineral water on at least half of the intervention days.In addition, consumption of less than 1 liter per day on more than 2 intervention days resulted in exclusion from the analysis.
In order to minimize the impact of circadian rhythmrelated fuctuations, all appointments for participants were set at approximately the same time within a consistent morning time frame, spanning from 6:00 to 10:00 a.m.Moreover, the study period began on Monday for all 2 Journal of Nutrition and Metabolism participants.Te examination days (t 0 , t 3 , t 28 ) were structured as follows: After completion of the current health status questionnaire, measurements of anthropometric characteristics were performed, followed by blood sampling.

Urine Sampling and Biochemical Measurements.
In order to assess the efect of water consumption on urinary acid-base parameters, participants were instructed to collect 24-hour urine samples on the day before each examination day and spontaneous urine samples on the examination days (t 0 , t 3 , and t 28 ).All participants were given detailed written instructions for collecting the 24-hour urine samples and were provided with preservative-free plastic containers (Sarstedt AG and Co. KG, Nümbrecht, Germany).Participants were instructed to begin the collection the day before their scheduled examination, beginning after voiding and excluding the frst morning urine.Te collected urine included the morning sample of the following day.Subjects were asked to record both the start and end times of the collection on the plastic containers.To check the completeness of the 24-hour urine collection, urinary creatinine excretion was measured.Upon arrival at the research institute, urine volume was measured and urine samples were immediately mixed.Tey were then divided into aliquots, and stored at −22 °C until analysis.One aliquot (1 urine tube) was stored at 5 °C until transferred to an external laboratory (LADR, Laboratory Network Hanover) for further analysis.After the arrival at the research institute, subjects were asked to collect another urine sample, which served as spontaneous urine.Urine pH was measured in both spontaneous urine and 24-hour urine samples using a pH meter (Mettler-Toledo, Giessen, Germany; Sartorius, Göttingen, Germany) at the Research Institute.Titratable acids (TA), ammonium (NH 4 + ), and bicarbonate (HCO 3 − ) were measured in 24hour urine specimens according to the 3-step titration method by Lüthy [17].Tese urine parameters were used to calculate renal net acid excretion (NAE) according to the following equation: NAE � TA − HCO Analysis System (Siemens Healthcare GmbH, Erlangen, Germany).Te remaining blood samples were stored at ≈5 °C and transferred to an accredited and certifed laboratory (LADR, Laboratory Network Hanover) for analysis.

Dietary Assessment and PRAL Calculation.
To evaluate any potential dietary changes that may have occurred during the intervention, the participants' dietary intake was assessed before (t 0 ) and at the end of the intervention period (t 28 ) using 3-day food records.In order to directly compare nutritional intake and urinary output, participants recorded the type and amount of all foods and beverages consumed over three consecutive days (Friday, Saturday, and Sunday) directly before the frst day of the examination (Monday).To take into account diferent mineralization of tap water, specifc tap water composition of every participant was entered in the database.Te composition of the participants' tap water was determined on the basis of the drinking water analysis of the water company that supplies the participants' particular area of residence.Nutrient intake was calculated from food and beverage intake (3 days).Moreover, the main fuid intake (beverages and soups) was evaluated at t 28 to assess fuid balance.Nutrient and energy intake were estimated using the software PRODI 6.12 Expert ® (Nutri- Science GmbH, Freiburg, Germany).

Journal of Nutrition and Metabolism 3
To evaluate the potential renal acid load (PRAL) of the nutrition, PRAL values of the food and beverages consumed were calculated using the formula proposed by Remer and Manz (PRAL 1995 ) [18].Te PRAL value of the mineral water is calculated according to the formula adapted by Wynn [14].
Despite this, the following assumption was used to calculate the amount of salt ingested: NaCl � 40% Na and 60% Cl.Terefore, the calculation equation for salt was: Cl (mg) × 100/60 � NaCl (mg).
2.9.Statistical Analysis.Sample size calculation was made for the primary aim of the study (NAE) considering a statistical power of 80% and a hypothesized efect size above 1.0.An a priori sample size calculation revealed that 36 participants (18 for each water group) would be required to detect a statistical diference of p < 0.05 in a parallel group design.
Distribution of all data was assessed using Shapiro-Wilk test and Q-Q-plots.Non-normally distributed data were transformed using the natural logarithm (ln) or square root (sqrt) transformation, whichever was more appropriate.For data that exhibited non-normal distribution after transformation, nonparametric testing was used.Data are presented as mean ± standard deviation or median and interquartile range (IQR), depending on the distribution.Te primary outcome measured is NAE.All other outcomes are secondary outcomes.
Diferences in baseline characteristics between the two intervention groups were analyzed using the unpaired t-test or Mann-Whitney U test.Nominal variables were tested using Fischer's exact test.
In terms of nutritional data, between-group diferences were analyzed using unpaired t-test or Mann-Whitney U test, while within-group diferences were analyzed using paired t-test or Wilcoxon test.Moreover, a two-way repeated measures ANOVA was computed to detect intervention efects (interactions between time and group), despite the partial absence of normal distribution and the partial presence of heteroscedasticity.According to Bortz [19], variance analyses are considered robust against violation of the normal distribution assumption with a sufciently large sample size (central limit theorem, n > 50), and inhomogeneous variances are not a problem with approximately equal group sizes.In addition, repeated-measures analyses of variance (rmANOVA) followed by Bonferroni correction for multiple testing (pairwise testing) were also performed between baseline (t 0 ), interim (t 3 ), and fnal (t 28 ) examinations to compare inner-group diferences.Twosided p values <0.05 were considered as statistically signifcant.All statistical analyses were performed using SPSS Software for Windows (Version 28.0.1.0;SPSS Inc., Chicago, IL, USA).

Study Population.
Te allocation of participants to the two intervention groups is shown in Figure 1.Ninety-four participants (67 women, 27 men) were included in this study, but 9 subjects were excluded from the evaluation due to dropouts and missing values (forgotten samples, illness or other personal reasons).One subject in the HBS group withdrew from the study due to gastrointestinal discomfort.No participant was excluded due to insufcient water consumption.Terefore, data from 85 participants (HBS group: n � 43, LBS group: n � 42) were analyzed.In terms of 24-hour urine data, there was one more missing data in the HBS group, resulting in 42 evaluable 24-hour urine samples.
General baseline characteristics of both intervention groups are presented in Table 2.At baseline, 59 women (69.4%) and 26 men (30.6%) aged 53 ± 10 years were included in the study.Tere were no signifcant diferences in anthropometric parameters between the two intervention groups, and these parameters did not change signifcantly from the beginning to the end of the intervention.Participants maintained their physical activity levels during the intervention period.

Dietary Intake of PRAL-Related Nutrients and Dietary
Acid Load.Intake of macronutrients as well as PRALrelated micronutrients is summarized in Table 3.In both intervention groups, the intake of macronutrients did not change signifcantly between baseline (t 0 ) and fnal examination (t 28 ), with no signifcant interaction between time and group.Moreover, there were no signifcant changes in micronutrient intake with respect to calcium, magnesium, and phosphorus intake in both intervention groups, with no signifcant interaction between time and group.On the contrary, the intake of sodium, chloride, and potassium difered between baseline (t 0 ) and fnal examination (t 28 ), depending on the intervention group.In the HBS group, sodium and chloride intake was signifcantly higher at the end of the study (t 28 ) compared to baseline (t 0 ), as a result of test water consumption (p < 0.001 and p � 0.001, respectively).In the LBS group, there were no signifcant changes in sodium intake, but chloride intake was signifcantly lower at the end of the intervention (t 28 ) (p � 0.025).Consequently, there was a signifcant change in salt consumption (NaCl) in both groups, but showing opposite directions.While the consumption of NaCl in the HBS group was signifcantly higher at the end of the intervention (t 28 ) compared to baseline (t 0 ) (p � 0.001), it was signifcantly lower in the LBS group at the end of the intervention (t 28 ) (p � 0.025).Furthermore, potassium intake in the LBS group was signifcantly lower at the end of the intervention (t 28 ) compared to baseline (t 0 ) (p � 0.002).
Moreover, PRAL values calculated from dietary protocols (PRAL 1995 ) showed a signifcant reduction of dietary acid load in the HBS water group (p < 0.001) and no change in the LBS water group (p � 0.200).

Bicarbonate and Salt
Intake from Mineral Water.Te two diferent mineral waters provided diferent amounts of minerals.While the HBS group consumed a high amount of additional bicarbonate, sodium, and chloride from the test water, the additional amounts of minerals from the consumed water in the LBS group were very low.While 4 Journal of Nutrition and Metabolism participants in the HBS group consumed an additional 6552 mg/d of bicarbonate, participants in the LBS group consumed an additional 342 mg/d of bicarbonate from the study water (1,500 mL).Te consumption of 1,500 mL HBS water led to an additional intake of 2.56 grams of sodium and 0.48 grams of chloride per day, while the consumption of LBS water led to an additional intake of 0.013 grams of sodium and 0.017 grams of chloride per day.Despite the diferent amounts of sodium and chloride, blood pressure did not change diferently between the two water groups (data published elsewhere) [20].With the exception of potassium and sulfate, which were present in higher concentrations in HBS mineral water (Table 1), no other constituent of the two mineral waters led to substantial diferent ingested amounts in minerals.

Efect of Mineral Water on Urine Volume and Urinary
Acid-Base Parameters.Urine volume did not change differently between the water groups (p � 0.176).Both groups  showed an increase in urine volume of about 400 mL within the frst 3 days of water consumption (p � 0.002 and p < 0.001 for HBS and LBS group, respectively) followed by stable urine volume until the end of the intervention (t 28 ) (p � 1.000 and p � 0.222 for HBS and LBS group, respectively).Consequently, long-term assessments also showed an increase in urine volume (p < 0.001 and p < 0.012 for HBS and LBS group, respectively).Tere was no signifcant diference in urine volume between groups at any time point (data not shown).
Daily mineral water consumption in the HBS group was 1,625 (214) mL of HBS mineral water, while LBS mineral water consumption in the LBS group was 1,643 (286) mL.At the end of the intervention, fuid intake and urinary excretion was balanced in both groups.Participants in the HBS group consumed 2,704 (1,100) mL of main fuids (beverages + soups), while the LBS group consumed 2,605 (822) mL.As shown in Table 4, urinary output was 2,687 (1,081) mL and 2,704 (1,021) mL, respectively.A comparison of fuid intake and urine volume showed no signifcant diferences within both groups (p HBS � 0.671, p LBS � 0.512).
As shown in Table 4, the efect of mineral water consumption on urinary acid-base parameters was substantial, with signifcant efects were primarily evident in the HBS group.
All urinary acid-base parameters showed signifcant interactions between time and group (all p < 0.001), indicating diferent efects of water consumption in the two intervention groups.While the consumption of HBS water resulted in signifcant changes in all urinary acid-base parameters (all p < 0.001), the consumption of LBS water signifcantly afected only a few parameters.
Specifcally, in the HBS group, urinary pH (24-hour urine and spontaneous urine) and HCO 3 − increased signifcantly within the frst 3 days of water consumption (all p < 0.001), followed by a stable high level until the end of the intervention (t 28 ) (p 24−h urin � 1.000, p spot urine � 0.097, p HCO3− �1.000).Consequently, long-term consumption of HBS water also resulted in signifcant increases in urine pH (24-hour urine and spontaneous urine) and HCO 3 − (all p < 0.001) of approximately the same magnitude as in the short-term period (Figure 2).In contrast, in the LBS group, HCO 3 − decreased slightly within the frst 3 days of water consumption (p � 0.009), followed by a stable period until the end of the intervention (t 28 ) (p � 0.805).However, long-term consumption of LBS water showed no signifcant efect (p � 0.092).
Moreover, TA and NH 4 + decreased signifcantly within the frst 3 days of water consumption in both intervention groups (HBS: both p < 0.001; LBS: p TA � p � 0.028, p NH4+ < 0.001), followed by almost stable levels until the end of the intervention (t 28 ) (HBS: p TA � 1.000, p NH4+ � 0.131; LBS: p TA � 0.129, p NH4+ � 0.830).However, the decrease in both TA and NH 4 + was signifcantly greater in the HBS group compared to the LBS group.In the HBS group, TA levels were almost negligible, while NH 4 + was reduced by half.Again, long-term consumption of HBS water resulted in signifcant reductions of TA and NH 4 + (both p < 0.001) of approximately the same magnitude as in the short-term period.In contrast, only NH 4 + was signifcantly reduced in the LBS group after long-term consumption of mineral water (p � 0.023).However, these declines were smaller than those in the HBS group.In addition, TA in the LBS group showed no signifcant changes over the long-term period of consumption (p � 0.520).
Finally, this resulted in a signifcant interaction between time and group for NAE (p < 0.001), indicating diferent efects of water consumption in the two water groups (Figure 3).Nevertheless, in both intervention groups, NAE signifcantly decreased within the frst 3 days of water consumption (p < 0.001 and p � 0.018 for HBS and LBS group, respectively).However, the reductions were more pronounced in the HBS group than in the LBS group.From the interim examination (t 3 ) until the end of the intervention (t 28 ), both groups showed a stable NAE (p � 0.875 and p � 0.377 for HBS and LBS group, respectively).Moreover, long-term consumption of both mineral waters showed a decline in NAE levels, which were only signifcant in the HBS group (p < 0.001 and p � 0.257 for HBS and LBS group, respectively).

Efect of Mineral Water on Blood Parameters of Acid-Base-Status.
As shown in Table 5, the efect of mineral water consumption on venous blood gas parameters was modest, with signifcant efects observed primarily in the HBS group.
In blood samples, pH and pCO 2 did not show a significant interaction between time and group (p � 0.145 and p � 0.303, respectively), indicating similar efects of water consumption in the two intervention groups.However, pH in the HBS group tended to increase during the initial 3 days of water consumption (p � 0.068) and signifcantly decreased from the interim examination (t 3 ) to the end of the intervention (t 28 ) (p < 0.001); no long-term changes were observed (p � 0.427).In the LBS group, only long-term changes were signifcant (p � 0.036), showing a decrease in pH.For pCO 2 , the HBS group showed a slight but nonsignifcant increase during the frst 3 days of water consumption (p � 0.807) and from the interim (t 3 ) examination to the end of the intervention (t 28 ) (p � 0.167), with a signifcant increase during the long-term consumption (p � 0.027).In the LBS group there were no signifcant changes.
Moreover, for HCO 3 − and BE signifcant time and group interactions were shown (p � 0.001 and p < 0.001, respectively), indicating diferent efects of water consumption.Diferences were signifcant only in the HBS group (both p < 0.001).Specifcally, HCO 3 − and BE levels significantly increased during the initial 3 days of water consumption (p � 0.004), followed by a slight, but nonsignifcant decrease of HCO 3 − (p � 1.000) and a significant decrease of BE (p � 0.015) until the end of the intervention (t 28 ).Long-term consumption of the HBS water also led to a signifcant increase in HCO 3 − and BE levels (p � 0.011 and p � 0.030, respectively).

Discussion
In this study, the daily consumption of 1500-2000 mL of bicarbonate-and sodium-rich mineral water led to signifcant positive changes in urinary and blood acid-base 6 Journal of Nutrition and Metabolism parameters.On the contrary, drinking the LBS water showed only slight efects.Te favorable outcomes of consuming high-bicarbonate-sodium (HBS) water were observed after a short-term consumption period (3 days) and maintained until the end of the study (28 days).Troughout the followup period, the majority of parameters in both blood and urine remained stable, maintaining similar levels to those observed after the initial short-term intervention.
Considering that the diet remained unchanged, except for the main beverages, it is reasonable to conclude that the observed diferences are not due to alterations in dietary intake.Instead, these diferences can be exclusively attributed to the impact of the mineral water.
Te long-term results of the present study are consistent with previous mineral water studies showing positive efects on acid-base balance after consumption of bicarbonate-rich mineral water [10,13,14,[21][22][23][24][25][26].Except one study, which evaluated efects on several acid-base parameters in blood and urine samples [10], most of the conducted mineral water studies concentrated on blood and/or urinary pH.Blood gas parameters and net acid excretion (NAE) are not commonly assessed in studies involving mineral water, with only three exceptions [10,13,25] having delved into this area.

Efects on Urine Volume and Urinary Acid-Base
Parameters.In the current study, urine volume increased signifcantly in both intervention groups.At the end of the study, both groups showed a neutral fuid balance, with no signifcant diferences between the main fuid intake and urinary fuid excretion.However, considering that the body eliminates water not solely through urine but also via skin, feces, and respiration [36], a slightly negative fuid balance could be assumed according to the presented data.Nevertheless, fuids may have been balanced, because measured fuid intake just covered beverages and fuid intake from soups.Additional fuid intake from fruits, legumes, and other foods was not evaluated, leading to a lower reported fuid intake than actually consumed.Moreover, there were no diferences in urine volume between the two study groups at all time points.Tis indicated that both study groups   underwent comparable dilution efects due to urine volume.Te fndings align with previous mineral water studies, where signifcant increases have been observed in response to specifc daily intakes ranging from 1,500 ml to 2,000 ml of mineral water [10,14,[22][23][24].With the exception of one study, the magnitudes of increase did not diverge between intervention groups within each study.In the mentioned study, the group consuming bicarbonate-rich mineral water displayed an elevation in urine volume during the initial 4 weeks, followed by a reduction in the magnitude of increase almost returning to baseline levels.Conversely, the control groups exhibited minor declines in volume during the initial 8 weeks, followed by a slight increase by the end of the intervention [23].In addition, only one study showed a reduction in urine volume after the consumption of alkalizing mineral water [13].

Journal of Nutrition and Metabolism
Furthermore, impacts on urinary acid-base parameters include changes in pH (24-hour urine and spontaneous urine), TA, HCO 3 − , and NH 4 + .Te latter three parameters can be summarized as urinary NAE, serving as an indicator for dietary acid load [37].As mentioned above, urinary parameters were strongly infuenced by the consumption of HBS water within this study.Te very high HCO 3 − intake through HBS water resulted in an excess of HCO 3 − being excreted via urine.Te bufering efect was substantial; TA became virtually undetectable.All efects were highly signifcant in the short-term and in the long-term analysis.For pH in 24-hour urine, these fndings are consistent with the results of several previous mineral water studies [10,13,14,[21][22][23][24][25][26].In the previous studies, pH in 24-hour urine increased by about 0.29 to 0.80 points (mean), whereas the consumption of HBS water resulted in an even more pronounced pH increase of about 1.19 points (mean).Concerning spontaneous urine, only a single prior mineral water study has examined the efect on pH, demonstrating comparatively smaller positive efects than those seen after HBS water consumption [10].Regarding acid-base parameters determined by titration, there are only a few previous studies evaluating the impact on these parameters in mineral water studies [10,14,22], all showing similar efects.In the largest study conducted with 129 healthy participants, three diferent bicarbonate-rich mineral waters with diferent PRAL values (MBMP water, HBLP water, and MBLP water) were tested against low-bicarbonate mineral water with a high PRAL value (LBHP water) [10].While MBMP water was characterized by a moderate HCO 3 − content and a medium PRAL value, HBLP water had a high HCO 3 − content and a low PRAL value, and MBPL water had a medium HCO 3 − content and a low PRAL value.In this study, the consumption of the bicarbonate-rich mineral waters with diferent PRAL values resulted in favorable efects on urinary TA, HCO 3 − , NH 4 + , and NAE compared to the low-bicarbonate mineral water with a high PRAL value.Nonetheless, some distinctions emerged among the three bicarbonate-rich mineral waters used in this study.Te positive changes in urinary parameters were more pronounced in the groups drinking bicarbonaterich mineral water with a low PRAL value (HBLP and MBLP water).Te researchers concluded that alkalinity (PRAL value in mineral water) has a greater infuence on acid-base parameters than HCO 3 − content [10].Under the same presumption, it was anticipated that the acid-base balance in the current study would be more profoundly afected by HBS water consumption in the current study.Tis was attributed to its higher alkalinity (PRAL −63.07), resulting from a high sodium content.In fact, although the efects on urinary parameters showed the same direction in both studies, the effects of HBS water consumption surpassed those exhibited by the waters investigated by Wasserfurth et al. [10].It is noteworthy that baseline NAE was lower than expected.Typically, individuals following an omnivorous Western diet display a NAE ranging between 70 and 100 mEq/ day [38].In contrast, in both intervention groups, NAE was about 27 mEq/day at baseline.Tere are two explanations for such a low NAE.Firstly, the study population may have been more health conscious than the general population.Tey may have consumed a more healthy and therefore less acidic diet than the average population.Lower NAE values have already been reported with a plant-based diet [39].Moreover, there are several studies that have reported similar baseline NAE values as reported in the current study following omnivorous diets [10,[40][41][42].Secondly, urinary bicarbonate, a component of NAE calculations, may have been somewhat overestimated by titration of HCO 3 − [17].Considering the initial low NAE levels and the substantial bufering impact on urinary acidbase parameters, the substantial reduction in NAE to negative values is not surprising.Te negative NAE is consistent with a previous study showing that certain diets can lead to negative values [37].
Despite the direct infuence of diet on NAE, several hormones are involved in the regulation of renal acid-base excretion.Higher levels of aldosterone, angiotensin II, endothelin, parathyroid hormone, and glucocorticoids are linked to higher renal acid excretion.Moreover, metabolic acidosis has been shown to increase hormonal activity [43].Terefore, the correction of the slightly low-grade metabolic acidosis may have involved a change in hormonal regulation leading to a change in TA, HCO 3 − , and NH 4 + excretion, possibly caused by the mineral water consumption.Nevertheless, this is speculative, as only aldosterone has been examined in the present study.As previously reported, the consumption of bicarbonate-and sodium-rich mineral water in this study resulted in a signifcant decrease in urinary aldosterone [20], which may have contributed to a lower NAE.

Efects on Venous Acid-Base Blood Gas Parameters.
Efects on acid-base blood gas parameters include pH, BE, HCO 3 − , and pCO 2 .Tis is the frst study showing benefcial efects of short-term and long-term consumption of mineral water very high in bicarbonate and high in sodium on venous blood gas parameters.
As mentioned above, long-term efects have already been demonstrated in previous studies [10,13,25] and could be supported by this study results.In line with our fndings, the study carried out by Wasserfurth et al. [10] similarly indicated no noteworthy shifts in venous blood pH after a 4week consumption of three bicarbonate-rich mineral waters possessing diferent PRAL values.Nevertheless, in the study presented here, pH values in blood tended to increase over the frst three intervention days.Tis may also have occurred in the Wasserfurth study, but it could not be determined due to the long-term design of the study.Moreover, the short-term results of the current study are similar to the results of a previous intervention study with an alkaline mineral water (pH 10).In that study, the consumption of mineral water resulted in a signifcant increase in blood pH after a rather short duration of 2 weeks [13].However, capillary blood was used.Although arterial blood analyses are required for the evaluation of respiratory conditions, metabolic disturbances can also be adequately measured in venous blood samples [44,45].Venous pH, bicarbonate, and base excess closely align with arterial values.However, the agreement between arterial and venous pCO 2 is inconsistent [44].Under normal physiological conditions, blood pH is kept in a narrow range between 7.35 and 7.45 (arterial pH) to maintain physiological processes in the body, mediated by intracellular and interstitial pH (4.8).Te varying efects of bicarbonate-rich mineral water on blood pH in the short term and long term may be the result of efcient bufering systems to maintain a stable blood pH.Although drinking mineral water can lead to a transient pH increase, bufering systems react sequentially to keep it within the normal range over the long term.
In addition, within the scope of this study, there was a noteworthy increase in HCO 3 − and BE during the initial 3day phase of mineral water consumption and these elevated concentrations remained until the end of the intervention.Tis illustrates that the impact of consuming mineral water on blood acid-base balance becomes apparent shortly after the frst consumption and continues to endure over a long

10
Journal of Nutrition and Metabolism period of time, without any habituation efects.Tese fndings align with the outcomes of the study carried out by Wasserfurth et al. [10].In that research, noteworthy impacts on HCO 3 − and BE were exclusively observed in the subset consuming mineral water with a moderately high HCO 3 − content and a low PRAL value (MBLP water, PRAL � −22.1).However, the group consuming bicarbonate-rich mineral water with a low PRAL (HBLP water, PRAL � −19.3) also displayed a tendency towards an increase in these parameters.Notably, no distinctions were evident for the group consuming bicarbonate-rich mineral water with a moderate PRAL value (MBMP water, PRAL � −10.8).Te authors conclude that alkalinity may be more important than bicarbonate in regulating acid-base balance in blood.Based on this assumption and considering that the PRAL value of the water consumed in this current study (HBS water, PRAL � −63.07) was even lower than the PRAL value of the HBLP water in Wasserfurth's study, a greater impact on blood gas parameters was anticipated in the present study.Nevertheless, the efect on HCO 3 − and BE was slightly lower with HBS water than the efect of low-PRAL mineral water in Wasserfurth's study (MBLP and HBLP group).
Although pCO 2 signifcantly increased from the beginning of the intervention to the end, changes were marginal following HBS water consumption.In contrast, the changes on pCO 2 in the MBHP group were more pronounced in the mineral water study by Wasserfurth et al. [10].

Efects on Blood
Pressure.Both mineral waters contained diferent amounts of sodium.Te evaluation of the changes in blood pressure due to mineral water consumption has been published elsewhere [20].For the sake of completeness, the blood pressure results are also mentioned here.Despite the diferent amounts of sodium, blood pressure did not change diferently between the two water groups.Furthermore, no diferences were found between men and women, or between younger (<50 years) and older (≥50 years) participants in either intervention group.Tis is in line with the results of other studies in which the consumption of bicarbonate-and sodium-rich mineral water did not lead to blood pressure changes [26,[46][47][48].

Implications for Human
Health.Urine pH of healthy subjects ranges usually between 5.8 and 6.8, depending on the consumed diet [49].It has been shown that vegans have a higher urine pH (6.7) than omnivores (6.2) [50].Tis is in line with our fndings in the LBS group, where urine pH was about 6 and did not change signifcantly over the intervention period.Tus, mineral water consumption did not modify acid-base balance in this group.Moreover, NAE was only slightly positive in this group, indicating a moderate acidic diet.Since included subjects were without impaired kidney function, it is reasonable to assume that the excess acid from the diet would have been excreted without negative efects on human health.However, a lower urine pH value, associated with a higher NAE, can promote the formation of certain kidney stones [51].In addition, higher NAE values are associated with infammatory processes in the kidneys [52].
Base supplements and bicarbonate-rich mineral waters are commonly used for the modulation of urine composition.Teir alkalizing efect has been proven in several studies until now [10,14,22].Terefore, the most obvious impact on human health relates to the modulation of kidney stone risk (calcium oxalate stones, uric acid stones) due to changes in urinary pH.While a low urine pH favors crystallization, an elevated urine pH counteracts crystallization [49,53].Moreover, bicarbonate-rich mineral water has been shown to improve urinary inhibitors for crystallization, mainly citrate and magnesium [49].Accordingly, the consumption of bicarbonate-rich mineral water has a positive efect on the risk of calcium oxalate stones and uric acid stones due to its urine alkalizing efect and changes in urine composition.
In addition, as demonstrated by the lowering of NAE, bicarbonate-rich mineral water has a positive impact on acid-base-balance [10].Terefore, the negative efects of an acidic diet can be mitigated by its intake.On the one hand, a high intake of acid-producing precursors has been shown to negatively afect bones [54].On the other hand, the intake of bicarbonate-rich mineral water has been shown to lower bone resorption [14,25,55].Tus, it could be assumed that the consumption of bicarbonate-rich mineral water has a positive efect on bones due to its improvement in acidbase balance.

Strengths and Limitations.
Within the context of this study, there are some strengths and limitations that should be mentioned.One limitation is the absence of blinding in the study design which may have an infuence on the study outcome.Since the taste of the water alone indicated the group assignment, blinding was not feasible.Moreover, the lack of meal standardization should be recognized as a possible source of infuence on the study outcomes.Te inclusion of such standardized meals would have been benefcial to increase the robustness of our results.Tis approach would have taken into account the diferent efects of diferent meals on urinary acid excretion, thus providing more reliable results.Furthermore, it would have been benefcial to include additional control groups consuming mineral water high in sodium alone.Tis would have allowed for a more precise evaluation of the individual effects of HCO 3 − and PRAL of the mineral water and improved the study's ability to make specifc statements regarding their respective impacts.
Conversely, it is important to acknowledge the strengths of this study.A strength is the sample size, surpassing previous studies that examined similar efects with smaller participant groups.Tis increased sample size enhances the study's statistical power, making it more likely to detect true efects and enabling more precise estimates of the observed efects.Furthermore, assessing efects at multiple time points provides the opportunity to capture progression over time and uncover potential habituation efects.Another advantage of the study is the extensive data collection on various parameters, including blood gas and urinary parameters, providing a comprehensive view of potential efects.

Conclusion
In conclusion, this study highlights the signifcant and benefcial efects of mineral water consumption, characterized by a very high bicarbonate and high sodium content (HBS water) content, on acid-base parameters.In particular, the consumption of 1,500 to 2,000 mL mineral water led to a signifcant reduction in urinary NAE, as refected by decreased levels of TA and NH 4 + , with a concomitant increase in urinary HCO 3 − excretion.A signifcant increase in urinary pH was also observed.Moreover, blood gas parameters were positively afected, although to a lesser extent.Taken together, these results emphasize that the consumption of mineral water with high HCO 3 − and sodium concentrations exerts a benefcial efect on the acid-base balance in humans.Tese efects became apparent shortly after the beginning of consumption and showed no signs of habituation.Moreover, no adverse efects on human health have been observed.
Further research should address the diferent efects on acid-base parameters of consuming mineral water with a very low PRAL and simultaneous low bicarbonate content compared to a bicarbonate-rich mineral water.Tis research is warranted to isolate the potential efects of the cooccurrence of high HCO 3 − content and low PRAL.
Data are shown as median (IQR).HBS � high bicarbonate and sodium, LBS � low bicarbonate and sodium.TA �

Table 1 :
Mineral content and PRAL of the test products.

Table 2 :
Baseline (t 0 ) characteristics of the study groups.

Table 3 :
Dietary energy and nutrient intake calculated from 3-day dietary records at the beginning (t 0 ) and at the end of the intervention (t 28 ) as well as absolute changes (∆).
Data are shown as median (IQR).HBS � high bicarbonate and sodium, LBS � low bicarbonate and sodium.PRAL � potential renal acid load.* Time × group interactions were analyzed using two-way repeated measures ANOVA.* * Diferences between t 0 and t 28 within each group were analyzed using paired t-Test and Wilcoxon-Test.Te bold values are signifcant at p <0.05.

Table 4 :
Urine analysis and fuid consumption (main fuids) at the beginning of the study (t NAE at the beginning of the study (t 0 ), at the interim examination (t 3 ), and at the end of the intervention (t 28 ).

Table 5 :
Blood gas analysis at the beginning of the study (t 0 ), at the interim examination (t 3 ), and at the end of the intervention (t 28 ) as well as absolute changes (∆).Data are shown as median (IQR).HBS � high bicarbonate and sodium, LBS � low bicarbonate and sodium.BE � base excess, HCO 3 − � bicarbonate in venous blood, pCO 2 � partial pressure of carbon dioxide in venous blood.* Time × group interactions were analyzed using two-way repeated measures ANOVA.Diferences over the intervention time within each group were assessed with repeated measurements ANOVA.a � sign.Diferences between t 0 and t 3 (short term), b � sign.Diferences between t 3 and t 28 (follow-up), c � sign.Diferences between t 0 and t 28 (long term).Te bold values are signifcant at p <0.05.