My First Ever Paper (yes, all 4311 words of it)

May 27, 2010 at 9:52 pm (med, school)

Investigating the Diuretic Effects of a 300 mg Dose of Caffeine as Compared to Other Interventions

Abstract

Objective: Caffeine is widely reported in the media to be a dehydration-promoting diuretic, and hence to be avoided as a source for fluid replenishment. In this study, we seek to investigate and analyse the relationship of various interventions on urine output, with a special focus on the diuretic potential of a 300 mg dose of caffeine.

Design: Randomized, non-blinded single centre study.

Setting: Two multi-disciplinary laboratories in Imperial College, London.

Subjects: Healthy undergraduate male and female medical students (n=279), with age range 19 – 25 years.

Interventions: Students were assigned to receive one of 7 different treatments:

  1. 1 litre water load
  2. 1 litre saline solution (0.9%) load
  3. 1 litre Diet Bolt (300 mg caffeine, 0.4% taurine) load
  4. 1 litre Bolt (300 mg caffeine, 0.4% taurine, 100 g glucose) load
  5. 1 litre water load with DDAVP injection (0.25 ml, 1 mcg)
  6. 1 litre water load with 20 minutes of exercise
  7. 1 litre dextrose solution (5%, 50 g dextrose) load

Methods and Main Results: 279 healthy undergraduates were allocated as evenly as possible into one of 28 subgroups. Each subgroup received one of the 7 different interventions listed above on one of the two days of the study. Urine flow rate and specific gravity were measured before and after each treatment at 20 min intervals over a 3-hour period. A 300 mg dose of caffeine was not found to cause any significant increase in diuresis, whereas a 1 litre fluid load was sufficient to induce a significant increase in diuresis, with a peak urine flow rate at 60 min. The diuretic response to 1 litre of fluid returned to baseline levels at 120 min in all interventions. A decrease in urine flow rate was observed in the Bolt, saline and exercise intervention groups when compared against the water control group. There was a greater decrease in urine flow rate for the DDAVP intervention. When dextrose solution was given, urine flow rate and specific gravity changes were similar to water. In all intervention groups, there was an initial drop in specific gravity of urine over the first 60 min, and it then became fairly constant after the 100 min mark.

Conclusions: A 300 mg dose of caffeine does not cause a significant increase in diuresis, whereas a 1 litre fluid load induces a significant increase in diuresis which was not maintained for longer than 120 min. Specific gravity was found to be fairly constant across the intervention groups, indicating that the body maintains homeostasis primarily through excretion of excess water. It is hoped that this study will clarify lingering misconceptions on the diuretic and dehydrating properties of caffeine intake at moderate doses.

Key Words: caffeine, diuretic, water-electrolyte balance, salt load, water load

Introduction

Caffeinated drinks have attracted negative press in many instances on account of their purportedly adverse effects as a dehydration-promoting diuretic. Top athletes are often recommended by sports scientists to refrain from caffeinated beverages as a choice of fluid replenishment.1 Such advice extends to the general public too, who are told to avoid caffeine in situations where fluid balance may be compromised.2 Because caffeinated beverages are so widely consumed all over the world, the debate on the diuretic and possibly dehydrating properties of caffeine has been on-going over the past decade.

The physiological reasoning established for the diuretic properties of caffeine is due to its role as a non-selective adenosine receptor antagonist. Human studies have demonstrated that adenosine A1-receptor antagonists increase diuresis via the direct inhibition of tubular sodium absorption and tubuloglomerular feedback.3 Other studies involving knockout mice have also demonstrated that intact adenosine A1 receptors are required for caffeine-induced inhibition of renal sodium and fluid reabsorption,4 thus supporting the concept that antagonism of adenosine A1 receptors mediates these diuretic responses.

Previous studies conducted thus far have shown the diuretic effects of caffeine to only be significant upon ingestion of acute doses above 300 mg, and in individuals who have been deprived of caffeine for a period of weeks. 5-7 This is atypical of both caffeine dosage and consumption pattern in the general population, where average daily intake amounts to about 250 mg of caffeine.8

We hypothesized that 300 mg of caffeine would have a minimal effect on diuresis when compared against a water control group.  With these facts in mind, an experiment was designed to test the effects that a 300 mg dose of caffeine and other interventions would have on urine output. The dosage of caffeine to be administered (300 mg) was arrived at by approximating a figure similar to the 24-hour caffeine intake8 in an average person. The study was conducted in a manner that would not place subjects under artificial experimental conditions; i.e. there were no restrictions on dietary intake or consumption of caffeine over the period leading up to the study. These measures were adopted to reflect a typical scenario of caffeine consumption, and hence yield data more applicable to free-living individuals. Changes in diuresis are reflected by an increase or decrease in urine flow rate and specific gravity. The primary objective of the study was to test whether a 300 mg dose of caffeine has a diuretic effect, with the secondary objective being to investigate the diuretic effects of other fluids and interventions.

Materials and Methods

Study population

The study was conducted in 2 multidisciplinary laboratories in Imperial College, London, on the 17th and 18th of May 2010. A total of 279 healthy undergraduate male and female students were recruited for this experiment after an invitation from the supervising professor. 259 students completed the intervention, with 20 subjects dropping out because of gastrointestinal complaints or unwillingness to complete the intervention due to palatability issues.

Randomization process and intervention

Students were randomised and placed into 4 groups, A, B, C and D. The 4 groups were subdivided into 7 groups (ie A1-A7), giving a total of 28 subgroups. Each subgroup was divided into one of seven different intervention groups in a non-blinded fashion as shown in Table 1.

Table 1. Allocation of subgroups

Intervention Subgroups assigned Total number of subjects

(number who did not complete intervention)

  1. 1 litre water load as the control experiment
A1, B1, C1, D1 47 (0)
  1. 1 litre saline solution (0.9%) load
A2, B2, C2, D2 43 (15)
  1. 1 litre Diet Bolt (300 mg caffeine, 0.4% taurine) load
A3, B3, C3, D3 35 (0)
  1. 1 litre Bolt (300 mg caffeine, 0.4% taurine, 100 g glucose) load
A4, B4, C4, D4 36 (1)
  1. 1 litre water load with DDAVP injection (0.25 ml, 1 mcg)
A5, B5, C5, D5 40 (1)
  1. 1 litre water load with 20 minutes of exercise
A6, B6, C6, D6 38 (1)
  1. 1 litre dextrose solution (5%, 50 g dextrose) load
A7, B7, C7, D7 40 (2)

The essence of the interventions was to investigate their effect on urine output. Groups A and B received their intervention on 17th May, while groups C and D received their intervention on 18th May. All groups were followed over a period of 3-hours on the day of their intervention. The author of this paper was included amongst the subjects and placed in group C3.

A week before the experiment, the following instructions were given: to minimize alcohol intake after 6pm the night before, have a light early breakfast with up to 250 ml of fluid, and to avoid all caffeinated beverages on the day of the experiment. No restrictions were placed on consumption of caffeine in the period leading up to the experiment for the reason discussed in the introduction. Students were further instructed to take note of what was consumed for breakfast, and to take note of the time when they first passed urine.

On the day of their intervention, students were told to report at 9am to one of two multidisciplinary laboratories, where urine samples were collected before the start of the experiment. They were then assigned 1 litre of water, saline solution, dextrose solution, Bolt or Diet Bolt to drink within 10 minutes, according to their designated subgroups (Table 1). Of the 12 subgroups assigned to drink 1 litre of water, 4 subgroups had to exercise (running around the laboratory block) for 20 minutes an hour after drinking the water, another 4 subgroups were administered a DDAVP injection an hour after drinking the water, and there was no further intervention in the remaining 4 subgroups. All interventions were self-administered, with the exception of the DDAVP intervention that was administered by injection into a peripheral vein by a trained technician. In the saline intervention, students were allowed to add a small amount of fruit squash (non sugar-free) to the saline fluids to make it more palatable.

After these interventions, students were instructed to pass urine every 20 minutes. The participants kept a 3-hour log, in which they recorded the voided volumes and specific gravity of urine at baseline and every 20 minutes there after. Ten results were obtained from each individual in all, including the baseline measurement. The volume (ml) and specific gravity of each urine sample were measured using standard measuring cylinders and a dipstick test respectively. Urine flow rate (ml/min) at each time interval was calculated by dividing urine volume at that interval by the time elapsed between intervals. In individuals who were unable to pass urine at specific intervals, the flow rate for that interval is assumed to be 0 ml/min.

Statistical analysis

The mean urine flow rate and mean specific gravity of urine samples were calculated from data submitted by the 4 subgroups within each of the 7 intervention groups. This is presented as graphs with standard error shown under the results below.

Considering the sample size in each group and non-Gaussian distribution of the observations, a t-test was used to compare the differences between or within the intervention groups.

Urine peak flow rate was observed at approximately 60 min in all intervention groups. Thus in order to draw conclusions about the diuretic effect of 1 litre of fluid, the peak flow at 60 min in the water load group was compared against urine flow rate at 20 min to see if this increase was significant.

To test our hypothesis, urine flow rate for the Diet Bolt intervention group was compared against the water control group at peak flow at 60 min.

A two-tailed P value of less than 0.05 was considered statistically significant.

Ethical issues

All participants had given full and informed consent to be included in the study.

Results

Water intervention (control)

There was an increase in urine flow rate from 20 to 60 min. Urine flow rate at 60 min was significantly higher [3.4 (2.4-4.3461) ml vs. 9.1 (7.9-10.3) ml, p<0.0001] than the flow rate at 20 min (Figure 1A). Urine flow rate decreased from 80 to 180 min, with no diuretic response after the 120 min mark when compared to baseline urine flow rate. Specific gravity decreased from 20 to 80 min, and increased from 80 to 180 min (Figure 1B).

Saline intervention

There was an initial rise in urine flow rate over the first 60 min (Figure 2A). Urine flow rate was significantly lower after 60 min as compared to the water control group. Some trend towards an increment of specific gravity (Figure 2B) was seen in this group as compared to the water control group.

Diet Bolt intervention

There was no effect of Diet Bolt on diuresis and specific gravity for the initial 20 min. The diuretic effects of Diet Bolt appeared at 60 min maximally (Figure 3A). Even though urine flow rate appeared to be greater from 40 to 120 min compared to the water control group, this difference was not significant when tested for using an independent variable t-test, using a peak urine flow rate at 60 min as a reference point. Specific gravity of urine as compared against the water control group showed no significant difference (Figure 3B).

Personal urine flow rate and specific gravity results were largely similar to mean results collated from the Diet Bolt intervention group, with the main differences being a slower increase in urinary flow rate over the first 40 min (Figure 3C), and a higher peak urine flow rate at the 60 min interval. This might have been because the author found Diet Bolt to be unpalatable, and hence took longer than 10 minutes to consume the drink.

Bolt intervention

The increase in urine flow rate of Bolt over the first 80 min (Figure 4A) was lesser than the water control group. There was no significant change in urine specific gravity after a sharp drop at 20 min (Figure 4B).

Water load with DDAVP intervention

There was a sharp rise in urine flow rate at 60 min (Figure 5A). This was followed by a drop in urine flow rate at 80 min, and a low urine flow rate from 100 min onwards, while specific gravity increased continuously after the 100 min interval (Figure 5B).

Water load with exercise intervention

A rise was observed in urine flow rate at 60 min (Figure 6A). This was followed by a decrease in urine flow rate as compared to the water control group. Urine specific gravity (Figure 6B) was similar to the water control group.

Dextrose intervention

Urine flow rate (Figure 7A) and specific gravity changes (Figure 7B) in the saline intervention were similar to results of the water control group.

The Diet Bolt intervention had no significant difference at peak urine flow rate at 60 min as compared to the water control group [10.7 (9.2-12.2) ml vs. 9.1 (7.9-10.3) ml, p=0.086] (Figure 8A).

There was no significant difference in specific gravity between the Diet Bolt intervention and water control group using the 80 min trough value as a reference point. [1.0 (1.0-1.0) ml vs. 1.0 (1.0-1.0) ml, p=0.98] (Figure 8B).

Discussion

Two primary determinants of urine output volume are blood pressure and plasma osmolality, as detected by baroreceptors and osmoreceptors respectively. These two mechanisms directly influence the secretion of vasopressin (VP) and brain natriuretic peptide (BNP). A rise in blood pressure or plasma osmolality can potentially trigger a rise in VP and/or BNP levels, allowing for greater water excretion. VP governs water permeability in the late distal tubule and collecting duct through expression of aquaporin-2, hence regulating the excretion of water.

The release of VP can differ greatly amongst individuals according to basal secretion rate and osmotic threshold,9 and this could further explain why the author’s set of results differed slightly from the mean in the intervention group, and also account for differences within all the other intervention subgroups.

In this study the 60 min flow rate was significantly higher in all intervention groups, suggesting that 1 litre of fluid load is adequate to elicit a physiological diuretic response. Any diuretic response decreased at 120 min in all intervention groups, suggesting that 1 litre of fluid is not sufficient to maintain diuresis for longer than 120 min. Specific gravity remains fairly constant across the intervention groups, apart from the saline and DDAVP intervention group. This indicates that the body maintains homeostasis primarily through excretion of excess water.

Saline intervention

There was an initial rise in urine flow rate over the first 60 min. A possible mechanism for this change is that there was an increase in circulating fluid volume by the same amount as if a hypotonic solution was being consumed, triggering detection by baroreceptors and a resultant increase in VP secretion.

The saline intervention group had a significantly lower urine flow rate after 60 min as compared to the water control group. This is because it does not trigger a change in plasma osmolality as the saline solution is isotonic, resulting in only baroreceptors but not osmoreceptors being stimulated. Thus VP secreted is less than that expected than after administration of 1 litre of water, and this allows for lesser water reabsorption by the kidneys, resulting in a smaller urine flow rate as compared to the water control group.

There was some trend towards increment of specific gravity in this group as compared to the water control group. This is in accordance with the known finding that retention of sodium induces a reduction in water excretion without significant change in vasopressin, angiotensin II, or aldosterone clearance,10 thus resulting in greater solute excretion as compared to water.

Bolt intervention

The Bolt drink is hyperosmolar due to its glucose content, and thus the diuretic effect of Bolt was lesser when compared to the water control group. This is because hyperosmolar fluids cause an increase in plasma osmolality that is picked up by osmoreceptors, triggering an increase in VP secretion. This causes more water to be reabsorbed from the renal tubular fluid, resulting in decreased urine excretion.

Water load with DDAVP intervention

There was a sharp rise in urine flow rate at 60 min as the initial water load caused an increase in blood volume and a decrease in plasma osmolality. This was picked up by baroreceptors and osmoreceptors, resulting in a decreased VP secretion and hence greater urine output to remove the excess water via the mechanisms discussed above.

Urine flow rate decreased and specific gravity increased sharply after a 100 min as the kidney concentrates urine in response to DDAVP. This phenomenon is due to the physiological effects produced by DDAVP, which is essentially a VP agonist.

Water load with exercise intervention

Similar to the DDAVP intervention group, there was a sharp rise in urine flow rate at 60 min due to the initial water load by the same mechanism as described above. This was followed by a decrease in urine flow rate after the 60 min interval as compared to the water control group, due to renal blood flow being diverted to muscles during exercise and an exercise-induced increase in VP and BNP secretion11.

Dextrose solution intervention

Urine flow rate and specific gravity changes were similar to results of the water control group, probably due to the small amount of dextrose arriving from the hepatic portal vein being fully converted to glycogen and removed from the blood by the liver. As resulting plasma osmolality is similar to water, any effects it had on urine flow rate and specific gravity was caused by the resultant increase in blood volume in a manner that is similar to water.

Diet Bolt intervention

As there was no difference in urine flow rate and specific gravity when compared to the water control despite an increasing trend observed in both groups, hence it is established that a 300 mg dose of caffeine does not have any effect on diuresis. This probably holds true for doses smaller than 300 mg as well. The increase in diuresis observed could be due solely to the effects of an increase in blood volume and reduction in plasma osmolality. Our finding is consistent with those found in other studies.1 12 13 14

A review conducted in 200815 of double-blind, placebo-controlled studies published over the past 15 years looking at the effect of caffeine on hydration concluded that caffeine intake of up to 400 mg per day did not produce dehydration, even in subjects undergoing exercise testing. The reviewer also found that the range of caffeine intake that could maximise benefit and minimise risk in relation to hydration is 38 to 400 mg per day. As noted in the introduction, average daily consumption of caffeine in the general population is closer to 250 mg8. In the UK the average daily consumption of caffeine among adults is about 200 mg/day (Figure 9), which falls well within this range. This suggests that associated risk from dehydration is extremely unlikely.

Figure 9. Daily caffeine consumption in different countries

(Reference: The World of Caffeine. Weinberg et al, 2001.)

Tolerance to caffeine rapidly develops16 so that any diuretic effect is diminished in people who regularly drink caffeinated beverages. Other studies have gone on to show that moderate intake of caffeinated drinks (250 – 300 mg caffeine/day) can contribute towards daily fluid intake and hydration17 18, and therefore advising seasoned caffeine drinkers to change their habitual drinking patterns to avoid its diuretic effects could actually lead to a decrease in overall fluid intake if this difference is not made up for by increasing intake of other fluids.

Limitations and suggestions for improvement

It should be noted that this study had several limitations. First and foremost, urine samples were collected over a relatively short 3-hour period after the various interventions. It has been shown in previous studies17 that in order to make valid conclusions about any diuretic effect, the time period over which urine output is collected is a significant factor. This is especially so with regard to caffeine intake as absorption from the gastrointestinal tract is rapid and virtually complete 45 minutes after ingestion, with the peak plasma caffeine concentration reached 15-120 minutes after ingestion.8 Studies were further able to demonstrate that the slight increase in urine output in response to caffeine disappeared within 4 hours, and this was compensated for by a decrease in output later in the day12. Thus, because most of caffeine’s diuretic effect is seen over a relatively short period after ingestion, especially after a single large dose, short-term urine collections can lead to misinterpretations about caffeine’s overall effect on urine output. Ideally, the study could have been designed to collect urine output over a 24-hour period, but would not have proven feasible in this instance as limited access to the laboratories was allowed.

Secondly, a confounding factor exists in that the average daily intake of caffeine and hence caffeine tolerance was not standardized among subjects. It has been shown that the body develops a tolerance to caffeine after three to five days of regular use,16 and this would have implications on any results collected.

Lastly, there was no attempt to standardize the initial hydration level in subjects before the start of the experiment, and this factor probably introduced the greatest error into results.

Without question, a metabolic ward setting would have yielded greater control, but these limitations to the study arose mainly because the objective was to obtain data that would be more applicable to free-living individuals. A follow-up study using change in body weight as a marker of fluid loss in subjects might prove to be of use in any future research investigating the dehydrating potential of caffeine. Estimation of plasma osmolality, which was not feasible in this study, might also yield better insight into drawn conclusions from future studies.

Conclusion

There is much evidence linking high experimental doses of caffeine with increased urine output, but this study has shown that a 300 mg dose of caffeine does not have any significant diuretic effect. The results from this study are more applicable to free-living individuals than previous studies, as it was conducted using a dosage that approximated the typical 24-hour caffeine intake of an average person, and in a manner mirroring that of a person who is not placed under stringent experimental conditions. It is hoped that lingering misconceptions on the diuretic and dehydrating properties of caffeine intake at moderate doses will be put to rest with the publication of this study.

References

1. Maughan RJ, Griffin J. Caffeine ingestion and fluid balance: a review. J Hum Nutr Diet 2003; Dec;16(6):411-20.

2. In-flight health. Available at: http://www.bbc.co.uk/health/treatments/travel/before_flighthealth.shtml. Accessed May 21, 2010.

3. Modlinger PS, Welch WJ. Adenosine A1 receptor antagonists and the kidney. Curr Opin Nephrol Hypertens 2003; Sep;12(5):497-502.

4. Rieg T, Steigele H, Schnermann J, Richter K, Osswald H, Vallon V. Requirement of intact adenosine A1 receptors for the diuretic and natriuretic action of the methylxanthines theophylline and caffeine. J Pharmacol Exp Ther 2005; Apr;313(1):403-9.

5. Passmore AP, Kondowe GB, Johnston GD. Renal and cardiovascular effects of caffeine: a dose-response study. Clin Sci (Lond) 1987; Jun;72(6):749-56.

6. Neuhauser-Berthold, Beine S, Verwied SC, Luhrmann PM. Coffee consumption and total body water homeostasis as measured by fluid balance and bioelectrical impedance analysis. Ann Nutr Metab 1997;41(1):29-36.

7. Robertson D, Frolich JC, Carr RK, Watson JT, Hollifield JW, Shand DG, et al. Effects of caffeine on plasma renin activity, catecholamines and blood pressure. N Engl J Med 1978; Jan 26;298(4):181-6.

8. Barone JJ, Roberts HR. Caffeine consumption. Food Chem Toxicol 1996; Jan;34(1):119-29.

9. Baylis PH. Osmoregulation and control of vasopressin secretion in healthy humans. Am J Physiol 1987; Nov;253(5 Pt 2):R671-8.

10. Os I, Aakesson I, Enger E. Plasma vasopressin in hereditary cranial diabetes insipidus. Acta Med Scand 1985;217(4):429-34.

11. Geny B, Charloux A, Lampert E, Lonsdorfer J, Haberey P, Piquard F. Enhanced brain natriuretic peptide response to peak exercise in heart transplant recipients. J Appl Physiol 1998; Dec;85(6):2270-6.

12. Grandjean AC, Reimers KJ, Bannick KE, Haven MC. The effect of caffeinated, non-caffeinated, caloric and non-caloric beverages on hydration. J Am Coll Nutr 2000; Oct;19(5):591-600.

13. Armstrong LE, Casa DJ, Maresh CM, Ganio MS. Caffeine, fluid-electrolyte balance, temperature regulation, and exercise-heat tolerance. Exerc Sport Sci Rev 2007; Jul;35(3):135-40.

14. Armstrong LE, Pumerantz AC, Roti MW, Judelson DA, Watson G, Dias JC, et al. Fluid, electrolyte, and renal indices of hydration during 11 days of controlled caffeine consumption. Int J Sport Nutr Exerc Metab 2005; Jun;15(3):252-65.

15. Ruxton, C.H.S. The impact of caffeine on mood, cognitive function, performance and hydration: a review of benefits and risks. Nutr Bulletin 2008; Mar;33(1):15-25.

16. Robertson D, Wade D, Workman R, Woosley RL, Oates JA. Tolerance to the humoral and hemodynamic effects of caffeine in man. J Clin Invest 1981; Apr;67(4):1111-7.

17. Ganio MS, Casa DJ, Armstrong LE, Maresh CM. Evidence-based approach to lingering hydration questions. Clin Sports Med 2007; Jan;26(1):1-16.

18. Popkin BM, Armstrong LE, Bray GM, Caballero B, Frei B, Willett WC. A new proposed guidance system for beverage consumption in the United States. Am J Clin Nutr 2006; Mar;83(3):529-42.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: