Secondary Logo

Journal Logo

Original article

Effects of a carbohydrate-electrolyte beverage on blood viscosity after dehydration in healthy adults

CHANG, Cui-qing; CHEN, Yan-bo; CHEN, Zhi-min; ZHANG, Lan-tao

Author Information
doi: 10.3760/cma.j.issn.0366-6999.2010.22.009
  • Free


Blood viscosity is determined by the hematocrit, erythrocyte aggregation, erythrocyte flexibility, platelet aggregation and plasma viscosity.1–4 Blood viscosity is higher in patients with cardiovascular diseases such as myocardial infarction, hypertension and peripheral vascular disease,5–7 and cerebral blood flow is suppressed in people with high blood viscosity.8,9 High blood viscosity may cause ischemic heart disease and stroke,10 and is associated with a risk of venous thrombosis.11,12 In order to prevent these viscosity related problems, it is important to avoid increases in blood viscosity, and the measurement of blood viscosity is included as a regular clinical test item for annual health examination at many hospital in China.

Body water losses due to perspiration and sweating decrease plasma volume and increase hematocrit, hemoglobin concentration, total protein concentration,13 the number of blood platelets and fibrinogen concentration,14 and cause an increase in blood viscosity, especially in a warm or hot environment.15–17 An increase in blood viscosity by sweat-induced dehydration can result in thrombus formation in the coronary arteries and cerebral vessels, and lead to sudden death in heat.14 Therefore, the prevention and quick recovery from sweat-induced dehydration is required to prevent thrombus formation and reduce the risk of sudden death from hemoconcentration and an increase in blood viscosity.

Carbohydrate-electrolyte beverages (CEs) have been reported to be more effective for retaining water in the body and preventing hemoconcentration than plain water.18,19 Plain water ingestion results in a fall of plasma electrolyte concentration and osmolality, which induces water diuresis and leads to voluntary dehydration.20,21 The addition of electrolytes to the ingested fluid maintains plasma electrolyte concentration and osmolality, and water diuresis is avoided. Drinking plain water after sweating also quenches thirst because electrolytes are lost into sweat, but water alone does not replenish lost fluid and causes dilution of blood and urine.20,21 This suggests that the ingestion of CEs after dehydration would be better than water for maintaining blood volume and recovery from hemoconcentration and increased blood viscosity.

The purpose of the present study was to evaluate whether CEs could attenuate the increase in blood viscosity due to hemoconcentration following exercise induced dehydration. The effectiveness of tea, typically consumed in China, and water was also determined.



This study was approved by Peking University Third Hospital Medical Ethical Committee. Ten healthy men were recruited from university students. All subjects were given written and oral complete information about the research project, purposes, procedures and potential risks, and had signed the consent forms before participating in this study. Their mean age, body weight, height, body mass index (BMI) were (26.0±1.8) years, (172.9±6.6) cm, (67.9±8.0) kg, (22.7±1.5) kg/m2, respectively.

Experimental design and procedure

This study had a randomized crossover control design with at least seven days as a washout period between the three trials. The effectiveness of three beverages for rehydration and decrease of blood viscosity following exercise-induced dehydration was evaluated during a 3-hour rehydration period (REP) while the subjects sat at rest in a thermo-neutral environment (26.3°C, 48% of relative humidity). Dehydration of 2.2% of body weight was induced by loading moderate exercise in a hot environment (34.3°C, 63% of relative humidity).

The subjects ingested commercial CE, water or tea equal to their body weight loss. The test beverages were consumed three times during the first 20 minutes in REP. The subjects ingested a volume equal to the amount of 1% body weight immediately at the post exercise rest period (0 minute in REP), and then the remaining beverage was ingested in two portions at 10 and 20 minutes in REP. The CE contained 21.0 meq/L Na+,5.0 meq/L K+ and 1 meq/L Ca2+, 6.6 g/100 ml carbohydrate (POCARI SWEAT, Tianjin Otsuka Beverage Co., Ltd., China).

The subjects consumed the same meal between 5:00 pm and 6:00 pm on the evening prior to each trial. They had free access to 600 ml of water from dinner until their bedtime. The subjects reported to the laboratory on the day of the experimental trials between 8:00 am and 9:00 am after overnight fasting. The subjects drank 600 ml of water to ensure euhydration after emptying their bladder, and were kept at rest for 60 minutes (pre-dehydration period). During this period, a catheter was placed in the forearm vein for further venous samplings. The first blood sample was obtained at the end of the pre-dehydration period. The subjects voided and were weighed immediately before and after the exercise dehydration period. Subjects repeated 20-minute bicycle exercise with 10-minute rest periods until reaching 2.2% of the total body weight loss in a hot environment. Pedaling resistance was adjusted to give a heart rate of 148 beats/min (65% of their estimated maximum heart rate).

After the dehydration period, the subjects were moved to thermo-neutral conditions in the laboratory, where they allowed to change their clothes and to rest in the seated position. A second blood sample was obtained at 20 minutes after the dehydration period (0 minute in REP). Blood samples were collected at 0, 10, 20, 30, 60, and 120, 180 minutes followed by urine collection and nude body weight was measured at 0, 30, 60, 120 and 180 minutes of the REP. The first access to test fluid was after the first urine collection at 0 minute. Thereafter, half of the remaining beverage was ingested after blood collection at 10 and 20 minutes, respectively. The subjects were instructed to stretch their knees on the chair for 10 minutes before every blood collection and to avoid sitting deeply in the chair during the REP period because a long-lasting seated position can lead to venous stasis induced hypovolemia. The inserted catheter was flushed with a small volume of heparinized saline after each blood collection. The experimental schedule is schematically presented in Figure 1.

Figure 1.
Figure 1.:
Experimental protocol on the test day. Body weight was decreased by 2.2% induced by bicycle exercise in a warm environment during the dehydration period. Total test beverage intake volume was equal to the body weight loss during the dehydration period. The beverage was consumed by three times, first time was at 0 minute with a volume equal to 1% of body weight before dehydration, then at 10 and 20 minutes with half of the remaining volume, respectively. Pre: Pre-dehydration period (60 minutes). Post: Post-dehydration period (20 minutes).


Body weight was measured with a flat scale with a 10 g minimum scale (HW-100KGL, A&D Co., Ltd., Japan). Urine volume collected in a bottle was measured with 1 g minimum scale. Heparinized whole blood was used for blood viscosity measurement with a rotational viscometer (SA-6000, Success Co., Ltd, China) at a shear rate of 200, 30, 5 and 1 per second. Viscosity was measured within 30 minutes after blood collection. Hematocrit and whole blood hemoglobin were determined using EDTA-treated blood with the capillary-centrifuge method and the cyan-methemoglobin method, respectively. The values were recorded as an average of duplicate measurements. Electrolytes and creatinine of serum and urine were measured with an automatic biochemical analyzer (AU-400, Olympus Co., Japan). The plasma and urine osmolalities were measured with a freezing point osmometer (Model 210, Fiske Micro-Osmometer, Advanced Inc., USA).


The retention rate of fluid consumed (FR) was calculated using the following formula: FR=100×(FI-cumulative urine volume)/FI, where FI is fluid intake volume. The percent change in plasma volume from pre-rehydration was calculated from the hematocrit and hemoglobin.22 The glomerular filtration rate (GFR) was indicated as the creatinine clearance (CCre) calculated by the following formula: CCre=UCre×V/BCre, where V is urine flow rate indicated as the urine volume per collection time, UCre and BCre were creatinine concentration in urine and serum, respectively. The free water clearance (CH2O) was calculated by the following formula: CH2O=V-(Uosm/Bosm) ×V, where Uosm and Bosm are osmolality in urine and plasma, respectively. The fractional excretion of sodium (FENa) was calculated by the following formula: FENa=100×(UNa×BCre)/(BNa×UCre), where UNa and BNa are sodium in urine and plasma, respectively. When urine was not excreted, then the GFR, CH2O and FENa were treated as a blank.

Statistical analysis

The time courses of the blood parameters were indicated as differences from the pre-rehydration values. The paired t-test was used to evaluate the significant differences at each point between CE and water. The same method was used for statistical analysis between tea and other beverages using the statistical package for Stat View, Version 5.0 software program (SAS, Inc., Cary, NC, USA) program and statistical significance was determined when P <0.05. Results are presented as the mean ± standard error of mean (SEM).



Total body water loss averaged (2.17±0.03)% for CE, (2.18±0.03)% for water and (2.16±0.03)% for tea during the dehydration period, showing no significant differences between the three trials with different test beverages. Therefore, the total ingestion volume did not differ between the three test beverages.

Blood viscosity

The blood viscosity was determined at 4 shear rates and the representative value at a shear rate of 5/s was used for analysis. The baseline value for blood viscosity was (7.41±0.27) mPa·s for CE, (7.14±0.15) mPa·s for water and (7.61±0.35) mPa·s for tea before dehydration. These values increased by (1.82±0.18) mPa·s for CE, (1.50±0.27) mPa·s for water and (1.55±0.23) mPa·s for tea after dehydration, showing no significant differences between the test beverages before and after dehydration. Blood viscosity in all test beverage trials began to decrease at 10 minutes in REP, and reached a plateau value at 30 minutes in water and tea, but CE remained decreased until 60 minutes and continuously kept in a lower level than water and tea during the following hours of REP. It showed significant differences between CE and water or tea at 60 minutes (P <0.05). Blood viscosity measured at other shear rates showed similar changes and representative values at a shear rate of 5/sec are shown in Figure 2.

Figure 2.
Figure 2.:
Changes in blood viscosity (at shear rate 5/s) during the rehydration period. Values are the mean±SEM. * P <0.05 vs. water. P <0.05 vs. tea by the paired t-test. Black square: carbohydrate-electrolyte beverage. Open circle: water. Gray triangle: tea.

Hematocrit and plasma volume

Hematocrit was (43.0±0.7)%, (42.8±0.6)% and (43.2±0.8)% before dehydration, and increased by (2.4±0.2)%, (2.1±0.3)% and (2.7±0.2)% after dehydration in CE, water, tea trials, respectively. Plasma volume decreased by (11.0±1.0)%, (10.3±1.1)% and (11.5±0.7)% after dehydration from the base line in CE, water and tea, respectively. There were no significant differences in the hematocrit and plasma volume between the test beverages.

In REP, hematocrit started to decrease at 10 minutes for all test beverages and tended to decrease further in water and in CE other than in tea during 60 minutes. For CE, a lower hematocrit level was sustained during the next 2 hours (60 to 180 minutes) compared with water. Between 60 to 180 minutes, the hematocrit was significantly higher in tea than CE (P <0.01) and water (P <0.05), respectively.

The percent changes in plasma volume started to elevate at 10 minutes in REP in all test beverages. After 60 minutes, the plasma volume remained higher until 180 minutes in CE than water, and a significant difference was observed between CE and water at 120 minutes (P <0.05). The increase of plasma volume in tea was lower than any other beverages, and the differences were significant during 60 to 180 minutes (Figure 3).

Figure 3.
Figure 3.:
Changes in plasma volume during the rehydration period. Values are the mean±SEM. * P <0.05 vs. water. P <0.05 vs. tea by the paired t-test. Black square: carbohydrate-electrolyte beverage. Open circle: water. Gray triangle: tea.

Urine volume and fluid retention rate

The mean urine volume amounted to only 4.3% (66 g) of the ingested volume in CE at 60 minutes in REP, significantly smaller than in water. The urine volume increased for all test beverages during the next hour (60 to 120 minutes), and it was significantly smaller in CE ((193±31) g) compared with water ((368±40) g) and tea ((358±40) g). During the last one hour in REP (120 to 180 minutes), the urine volume decreased (CE: (101±25) g, water: (142±28) g, tea: (135±28) g) showing no significant differences between any of the test beverages. The urine volume increased by 120 minutes in all test beverages, but the cumulative urine volumes were significantly less for CE compared with water and tea during the 0–180 minutes period. In addition, the fluid retention rate was significantly higher for CE ((77.0±3.9)g) compared with water ((61.2±3.4)%) and tea ((60.5±3.7)%) at 180 minutes (Figure 4).

Figure 4.
Figure 4.:
Cumulative urine output (A) and fluid retention (B) during the rehydration period. Values are the mean±SEM. * P <0.05 vs. water. P <0.05 vs. tea by the paired t-test. Black square: carbohydrate-electrolyte beverage. Open circle: water. Gray triangle: tea.
Figure 4.
Figure 4.:

Glomerular filtration rate, free water clearance, fractional excretion of sodium

The glomerular filtration rate was elevated after dehydration and returned to baseline between 0 to 30 minutes in REP for all of the test beverages. No significant differences were observed in the glomerular filtration rate between any of the test beverages at any of the time points. The free water clearance was lower for CE than water and tea at 120 minutes. The fractional excretion of sodium was 0.90%-1.17% during 60 to 180 minutes in tea, persistently higher than those in CE and water. No difference was observed for the fractional excretion of sodium between CE and water (Table).

Glomerular filtration rate, free water clearance and fractional excretion of sodium before and after dehydration and during rehydration

Correlation analysis of blood viscosity with hematocrit

A significant positive correlation was observed between the hematocrit and blood viscosity in all test beverages at all of the time points during REP (r=0.53–0.82, P <0.01).


In this experiment, the increased blood viscosity induced by dehydration was shown to decrease significantly by the CE, in comparison to water or tea. At the same time, when the CE was consumed, the fluid was well retained within the body and hemoconcentration was alleviated. Dehydration induced by sweating induces concentration of the blood, an increase in hematocrit and decreases in plasma volume. Dehydration is also associated with high blood viscosity because hematocrit influences blood viscosity. In the present study, dehydration and hemoconcentration were induced experimentally by exercise in a hot environment. In all of the trials, increases in hematocrit (2.1%-2.7%) and decreases in plasma volume (-10.3% to -11.5%) were observed after dehydration, showing similar trends to previous reports with similar levels of dehydration. 13 It is well known that dehydration with sweating induces not only hemoconcentration but also increases body temperature and heart rate and reduces exercise performance.23 Accordingly, in this experiment, we tried to determine the beverage appropriate for quick recovery from dehydration.

CEs maintained body fluid and restored plasma volume loss as seen in previous reports,18,19 and also showed a similar recovery from dehydration induced by sweating.20 This is because beverages containing electrolyte maintain plasma osmolality, and diuresis was less than with water ingestion.

The new finding of this study was that CE had a better effect on reducing blood viscosity than water and tea. The CE did not induce water diuresis and plasma volume was well maintained. Urine volume showed difference between CE and water during the 60 minutes after drinking, but it was smaller relative to the volume of consumption, and the difference was as small as 30 ml between CE and water. Urine volume of both CE and water increased in the next hour, and water had 180 ml more than CE. Water showed a higher free water clearance than CE between 60 to 120 minutes, indicating the urine in water was more diluted than that of the CE. Therefore, solutes including electrolytes left in the body play an important role in fluid retention, as reported by Nose et al.21 The plasma volume was increased higher in CE trial than in water trial. Blood viscosity tended to be lower with CE than with water. The strong correlation between hematocrit and blood viscosity suggests that fluid retained in the body reduced the hemoconcentration and blood viscosity.

In this study, blood sampling was performed at 10 minutes intervals up to 30 minutes after consumption of the test fluid to trace changes in the blood rheological parameters. The hemoconcentration started to decrease only 10 minutes after consumption and blood viscosity was also deceased. This suggested a rapid shift of ingested fluid from the intestine to the blood after consumption. However, blood viscosity reached a plateau after 20 minutes from the consumption of water, while in CE blood viscosity showed a decrease up to 60 minutes after the consumption. Similar changes were observed for hematocrit and plasma volume between CE and water, supporting the fact that CE was more effective for the reduction of hemoconcentration than water. The Na+/glucose co-transporter induces solute absorption and promotes fluid absorption in the epithelial cells of the intestinal tract by the osmotic gradient.24,25 Therefore, fluid absorption would be slower in water consumption, compared with CE. If an adequate volume of fluid is absorbed by the consumption of water, urine volume will increase with the dilution of plasma osmolality, but urine volume at 60-minute was smaller after the water consumption. Therefore, the ingested fluid remained in the intestines with water consumption, and caused the delayed tendency to recover from hemoconcentration.

We also conducted an exploratory study to examine the effect of common Chinese tea ingestion on blood viscosity. The results showed that tea had same reducing effect on blood viscosity as water during 20 to 120 minutes of REP, but not as well as water at 180 minutes obviously. The ingredients of tea, such as tannin, have been reported to delay stomach excretion. 26 Caffeine, another ingredient of tea, has a diuretic action. 27 However, tea showed no appreciable difference with water in reducing the hemoconcentration at 10–20 minutes after the consumption. The recovery from hemoconcentration was lower for tea than water after 30 minutes. Although the urine volume of tea was similar to water, its blood viscosity did not recover to the same level as water at 180 minutes. This may indicate that tea was excreted from the stomach more slowly than water after 20 minutes, and diuresis was increased. The glomerular filtration rate and free water clearance indicated that there was no difference in the effect on diuresis between tea and water. However, the fractional excretion of sodium of tea was higher than that of water. This can be explained by the limited renal tubular sodium reabsorption in tea. These results indicated that, when taken as a drink, tea was less effective in the recover from hemoconcentration than water. Nevertheless, tea had the same responses with reducing blood viscosity as water, the effectiveness may be contributed to some ingredients in tea with other pathways, which needs to be clarified.

Blood viscosity is higher in patients with cardiovascular diseases such as myocardial infarction, hypertension and peripheral vascular disease. High blood viscosity increases the risk of venous thrombosis. Blood viscosity is clinically tested as an indicator of disease progression in China. Sweat-induced water loss by exercise or bathing can increase blood viscosity and constitutes one of the risks of cerebral and cardiovascular infarction. In fact, the incidence of infarction is high during heat waves and dehydration after sweating leads to hemoconcentration and increased blood viscosity.

The present study compared the effect of CE and water and tea on the reduction of increased blood viscosity after dehydration. CE was highly effective to ameliorate hemoconcentration and the increase in blood viscosity, compared to water and tea. Tea was not found to be effective to correct hemoconcentration as good as water. These results suggest that CE is more useful to correct increased blood viscosity after dehydration than water and tea.


We sincerely thank research subjects who participated in the study, thank Dr. Taketoshi Morimoto, Professor Emeritus at Kyoto Prefectural University of Medicine and Mr. Masao Sakurai of Saga Nutraceuticals Research Institute, Otsuka Pharmaceutical Co., Ltd. for helping us with this research design, data analysis and review the manuscript. We thank DUAN Rui-ping for collecting blood samples in the test. The financial support of the Otsuka Pharmaceutical Co., Ltd. is also appreciated.


1. Chien S, Usami S, Taylor HM, Lundberg JL, Gregersen MI. Effects of hematocrit and plasma proteins on human blood rheology at low shear rates. J Appl Physiol 1966; 21: 81-87.
2. Dintenfass L. Blood viscosity, internal fluidity of the red cell, dynamic coagulation and the critical capillary radius as factors in the physiology and pathology of circulation and microcirculation. Med J Aust 1968; 1: 688-696.
3. Hathcock JJ. Flow effects on coagulation and thrombosis. Arterioscler Thromb Vasc Biol 2006; 26: 1729-1737.
4. Kensey KR. Rheology: an overlooked component of vascular disease. Clin Appl Thromb Hemost 2003; 9: 93-99.
5. Chien S. Blood rheology in myocardial infarction and hypertension. Biorheology 1986; 23: 633-653.
6. Danesh J, Collins R, Peto R, Lowe GD. Haematocrit, viscosity, erythrocyte sedimentation rate: meta-analyses of prospective studies of coronary heart disease. Eur Heart J 2000; 21: 515-520.
7. Tzoulaki I, Murray GD, Lee AJ, Rumley A, Lowe GD, Fowkes FG. Inflammatory, haemostatic, and rheological markers for incident peripheral arterial disease: Edinburgh Artery Study. Eur Heart J 2007; 28: 354-362.
8. Grotta J, Ackerman R, Correia J, Fallick G, Chang J. Whole blood viscosity parameters and cerebral blood flow. Stroke 1982; 13:296-301.
9. Wood JH, Kee DB Jr. Hemorheology of the cerebral circulation in stroke. Stroke 1985; 16: 765-772.
10. Lowe GD, Lee AJ, Rumley A, Price JF, Fowkes FG. Blood viscosity and risk of cardiovascular events: the Edinburgh Artery Study. Br J Haematol 1997; 96: 168-173.
11. Booth S, Chohan S, Curran JC, Karrison T, Schmitz A, Utset TO. Whole blood viscosity and arterial thrombotic events in patients with systemic lupus erythematosus. Arthritis Rheum 2007; 57: 845-850.
12. Dormandy JA, Edelman JB. High blood viscosity: an aetiological factor in venous thrombosis. Br J Surg 1973; 60: 187-190.
13. Nose H, Mack GW, Shi XR, Nadel ER. Shift in body fluid compartments after dehydration in humans. J Appl Physiol 1988; 65: 318-324.
14. Keatinge WR, Coleshaw SR, Easton JC, Cotter F, Mattock MB, Chelliah R. Increased platelet and red cell counts, blood viscosity, and plasma cholesterol levels during heat stress, and mortality from coronary and cerebral thrombosis. Am J Med 1986;81:795-800.
15. Brun JF, Connes P, Varlet-Marie E. Alterations of blood rheology during and after exercise are both consequences and modifiers of body's adaptation to muscular activity. Science & amp; Sports 2007; 22: 251-266.
16. Ernst E, Strziga P, Schmidlechner C, Magyarosy I. Sauna effects on hemorheology and other variables. Arch Phys Med Rehabil 1986; 67: 526-529.
17. Martin DG, Ferguson EW, Wigutoff S, Gawne T, Schoomaker EB. Blood viscosity responses to maximal exercise in endurance-trained and sedentary female subjects. J Appl Physiol 1985; 59: 348-353.
18. Doi T, Sakurai M, Hamada K, Matsumoto K, Yanagisawa K, Kikuchi N, et al. Plasma volume and blood viscosity during 4 h sitting in a dry environment: effect of prehydration. Aviat Space Environ Med 2004; 75: 500-504.
19. Hamada K, Doi T, Sakura M, Matsumoto K, Yanagisawa K, Suzuki T, et al. Effects of hydration on fluid balance and lower-extremity blood viscosity during long airplane flights. JAMA 2002; 287: 844-845.
20. Morimoto T, Miki K, Nose H, Yamada S, Hirakawa K, Matsubara C. Changes in body fluid volume and its composition during heavy sweating and effect of fluid and electrolyte replacement. Jpn J Biometeor 1981; 18: 31-39.
21. Nose H, Mack GW, Shi XR, Nadel ER. Role of osmolality and plasma volume during rehydration in humans. J Appl Physiol 1988; 65: 325-331.
22. Dill DB, Costill DL. Calculation of percentage changes in volumes of blood, plasma, and red cells in dehydration. J Appl Physiol 1974; 37: 247-248.
23. Sawka MN, Montain SJ, Latzka WA. Body fluid balance during exercise-heat exposure. In: Buskirk ER, Puhl SM, eds. Body fluid balance: exercise and sport. CRC Press, New York; 1996: 143-161.
24. Shi X, Summers RW, Schedl HP, Flanagan SW, Chang R, Gisolfi CV. Effects of carbohydrate type and concentration and solution osmolality on water absorption. Med Sci Sports Exerc 1995; 27: 1607-1615.
25. Turner JR. Show me the pathway! Regulation of paracellular permeability by Na+-glucose cotransport. Adv Drug Deliv Rev 2000; 41: 265-281.
26. Tani S, Yamazaki H, Muto N. Gastric emptying in chemical-administered and ulcerated rats. Yakugaku Zasshi 1983; 103: 225-230.
27. Del Coso J, Estevez E, Mora-Rodriguez R. Caffeine during exercise in the heat: thermoregulation and fluid-electrolyte balance. Med Sci Sports Exerc 2009; 41: 164-173.

carbohydrate-electrolyte beverages; blood viscosity; hemoconcentration; plasma volume; fluid retention

© 2010 Chinese Medical Association