J. Appl. Physiol.
88: 540–550, 2000.
NaHCO3 and KHCO3 ingestion rapidly increases
renal electrolyte excretion in humans
MICHAEL I. LINDINGER,1 THOMAS W. FRANKLIN,1 LARRY C. LANDS,2
PREBEN K. PEDERSEN,3 DONALD G. WELSH,1 AND GEORGE J. F. HEIGENHAUSER2
1Department of Human Biology and Nutritional Sciences, University of Guelph, Guelph N1G 2W1;
2Department of Medicine, McMaster University, Hamilton, Ontario, Canada L8N 3Z5; and
3Department of Sports Science and Physical Education, University of Odense,
DK 5230 Odense, Denmark
potassium bicarbonate; sodium bicarbonate; kidney; aldosterone; acid-base; strong ion difference; chloride; glomerular
filtration rate; urine alkalinization
IN ATTEMPTS TO IMPROVE short-term, high-intensity exercise performance, NaHCO3 loading has long been used
(19). There is, however, little information regarding the
time course and magnitude of acute renal responses in
humans who have ingested an amount of NaHCO3 that
may be considered to be of ergogenic benefit (minimum
The costs of publication of this article were defrayed in part by the
payment of page charges. The article must therefore be hereby
marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
solely to indicate this fact.
540
of 0.3 g/kg body mass). The chronic responses to ingested or infused bicarbonate solutions in clinical situations have been extensively documented (2, 9, 14, 23,
30, 37). Few studies, however, have examined the
effects of ingested KHCO3 (37, 38), and there do not
appear to be studies that have compared the early renal
responses to large, equimolar doses of NaHCO3 (sufficient to be of ergogenic benefit in humans, see Ref. 27)
and KHCO3 in humans. The physicochemical origins of
plasma fluid and ion disturbances to ingested Na1 and
K1 are expected to be different due to differences in
their distribution and cellular transport in the intestine, kidneys, muscles, and other tissues (27). The bulk
of Na1 absorbed from the intestinal tract remains in
the extracellular fluids (ECFs) and, if not fully excreted, will result in an increased ECF volume (ECFV);
on the other hand, K1 rapidly enters intracellular fluid
compartments (27).
Although acute effects of KHCO3 ingestion have not
been extensively studied in humans (27, 37, 38), a
generalized comparison of the responses to NaHCO3
and KHCO3 may be obtained from various studies.
Renal responses to ingested or infused NaHCO3 or
NaCl and KHCO3 or KCl include increased excretion of
the cation and water and a decreased reabsorption of
HCO2
3 (2, 30, 37). There are, however, some notable
differences among responses, depending on the cation
(Na1 vs. K1) and the accompanying anion (Cl2 or
HCO2
3 ). NaHCO3 loading, compared with NaCl loading,
results in increased renal excretion of K1 (UK1V) (14)
and Na1 (UNa1V) (22), whereas NaCl loading has no
effect on UK1V and Cl2 excretion (37). KCl loading,
compared with NaCl loading, results in an increased
glomerular filtration rate (GFR) (29), increased plasma
aldosterone (30), and increased renal UK1V and Cl2
excretion (24, 30, 37, 38). KCl loading is also associated
with decreases in plasma volume (PV) and ECFV (3, 37).
KHCO3 loading, compared with KCl loading, results in a
prolonged period of urine alkalinization associated with
2
increased HCO2
3 and Cl excretion and decreased ex1
cretion of NH1
(U
V)
and
titratable acid (TA) (37).
NH4
4
The main purposes of this paper are to describe and
interpret the acute renal contribution to the correction
of the volume, acid-base, and ion disturbances resulting from NaHCO3 and KHCO3 loading in humans. An
aim of this paper is to integrate the renal responses
8750-7587/00 $5.00 Copyright r 2000 the American Physiological Society
http://www.jap.org
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
Lindinger, Michael I., Thomas W. Franklin, Larry C.
Lands, Preben K. Pedersen, Donald G. Welsh, and
George J. F. Heigenhauser. NaHCO3 and KHCO3 ingestion
rapidly increases renal electrolyte excretion in humans. J.
Appl. Physiol. 88: 540–550, 2000.—This paper describes and
quantifies acute responses of the kidneys in correcting plasma
volume, acid-base, and ion disturbances resulting from
NaHCO3 and KHCO3 ingestion. Renal excretion of ions and
water was studied in five men after ingestion of 3.57 mmol/kg
body mass of sodium bicarbonate (NaHCO3) and, in a separate trial, potassium bicarbonate (KHCO3). Subjects had a
Foley catheter inserted into the bladder and indwelling
catheters placed into an antecubital vein and a brachial
artery. Blood and urine were sampled in the 30-min period
before, the 60-min period during, and the 210-min period
after ingestion of the solutions. NaHCO3 ingestion resulted in
a rapid, transient diuresis and natriuresis. Cumulative urine
output was 44 6 11% of ingested volume, resulting in a 555 6
119 ml increase in total body water at the end of the
experiment. The cumulative increase (above basal levels) in
renal Na1 excretion accounted for 24 6 2% of ingested Na1. In
the KHCO3 trial, arterial plasma K1 concentration rapidly
increased from 4.25 6 0.10 to a peak of 7.17 6 0.13 meq/l 140
min after the beginning of ingestion. This increase resulted in
a pronounced, transient diuresis, with cumulative urine
output at 270 min similar to the volume ingested, natriuresis,
and a pronounced kaliuresis that was maintained until the
end of the experiment. Cumulative (above basal) renal K1
excretion at 270 min accounted for 26 6 5% of ingested K1.
The kidneys were important in mediating rapid corrections of
substantial portions of the fluid and electrolyte disturbances
resulting from ingestion of KHCO3 and NaHCO3 solutions.
541
RENAL RESPONSES TO KHCO3 AND NAHCO3 LOADING
with the vascular and skeletal muscle responses described previously in these subjects (27). The hypothesis was tested that the kidneys play an important role
in the acute correction of fluid and ion disturbances
resulting from NaHCO3 and KHCO3 ingestion. It was
also hypothesized that in the NaHCO3 trial, compared
with the KHCO3 trial, the kidneys would excrete excess
water, solute, and base equivalents at a greater rate,
resulting in a more rapid correction of the fluid and ion
disturbance. This study also analyzed urine composition with respect to the independent variables of acidbase control in biological fluids, namely the strong ion
difference (SID) concentration ([SID]), the total concentration of weak acids and bases ([Atot]), and the concentration of CO2 (28, 35). These three variables determine
1
the measured concentrations of H1 and HCO2
3 ([H ]
and [HCO2
],
respectively),
and
analysis
of
changes
in
3
these variables provides insight as to the mechanisms
of acid-base control.
Five healthy men (age 29 6 2 yr, mass 80 6 5 kg)
participated in this study. Written informed consent was
obtained after the procedures and potential risks were fully
described to the subjects. The study was approved by the
University’s human ethics committee. A sixth subject was
treated for hyperkalemia (see Ref. 27); this individual’s data
were excluded from the experiment.
Experimental protocol. In the 24-h period before each trial,
subjects abstained from caffeine and alcohol. About 2 h before
their arrival at the laboratory, the subjects ate a light meal
(toasted bread and juice). All experiments began at ,8:00 AM
and consisted of ,1 h of preparation time and 5 h of data
collection.
A brachial artery and antecubital vein (opposite arms)
were catheterized percutaneously with a 20-gauge 1.25-in.long Teflon catheter (Angiocath, Becton Dickinson, Baxter,
Mississauga, ON) after the skin was infiltrated with 0.5 ml of
2% Xylocaine without epinephrine (Astra Pharma, Mississauga, ON). The patency of the catheter was maintained
using a slow saline drip (,200 µl/min). The urinary tract was
infiltrated aseptically with Xylo-Gel (Astra Pharma) and a
Foley urinary catheter (Baxter) inserted into the bladder. The
urinary analgesic Pyridium (Parke-Davis, Scarborough, ON)
was prescribed for 1–3 days following each experiment.
After insertion of the catheters, each subject was seated in
a comfortable chair for the remainder of the experiment.
During a 30-min baseline period, urine and blood samples
were obtained at 15-min intervals. At the end of this period
(time 0), subjects ingested 3.57 mmol/kg body mass of either
KHCO3 or NaHCO3 during the next 60 min. The ,920 ml of
solution ingested, which had an osmolarity of ,600 mosmol/l,
was flavored with Kool-Aid and sweetened with Nutrasweet.
The order of presenting the experimental treatments was
randomized for each trial, and trials were separated by at
least 2 wk to allow for normalization of hemoglobin concentration. The subjects were observed for a further 210 min in the
postingestion period. Urine drained continuously into a sealed
collection bag and was completely collected at 20-min intervals until 120 min and then at 30-min intervals until 270 min.
Measurements and analysis. Arterial and venous blood
sampling and analysis have been described (27). Arterial
plasma aldosterone concentration was determined by
RIA (Coat-A-Count TKAL1, Diagnostics Products, Los Angeles, CA).
[TA 2 HCO3],
meq/l 5
(EPBTV 3 Nb) 2 (Va 3 Na)
Vu
(1)
where EPBTV is the end-point base titration volume (liters),
or the volume of base added to reach the desired pH end point;
Nb and Na are the normality of NaOH and HNO3; Vu is the
volume of urine titrated (liters); and Va is the volume of HNO3
added (liters) to remove the HCO2
3 . Net acid excretion was
V).
calculated as UNa14 V 1 TA2HCO2
3 excretion (UTA2HCO2
3
Creatinine clearance was used as an estimate of GFR and
was calculated as previously described (36). In normal subjects, during the time of day when the study was conducted (8
AM to 1 PM), creatinine clearance reportedly exceeded GFR
by a nearly constant 12 6 2 ml/min (mean 6 SE, n 5 14; Ref.
36). Therefore, the small tubular secretion of creatinine is not
expected to affect the time course of the observed responses in
the present study. Ion excretion rates and ion fractional
excretions were calculated with standard equations (36).
Cumulative electrolyte excretion was determined by integration of ion excretion rates over time. Basal electrolyte
excretion was calculated on the basis of the preingestion
excretion rates. The difference between total cumulative and
cumulative basal excretion was referred to as ‘‘extra’’ and
represents the amount of ion excreted in excess of basal
levels.
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
METHODS
Urine volume was measured with graduated cylinders at
timed intervals for calculation of urine flow rate (UFR). Urine
pH was immediately measured (Brinkman Metrohm 632 pH
meter). Urine lactate and ammonium ([NH1
4 ]) concentrations
were measured by using enzymatic fluorometric techniques
(5) on 400-µl samples deproteinized in 800 µl of 6% perchloric
acid. Urine Pi concentration ([Pi]) was assayed by spectrophotometric analysis (Sigma kit 670, Sigma Chemical, St. Louis,
MO). Urine sodium ([Na1]), potassium ([K1]), and calcium
([Ca21]) concentrations were measured, after appropriate
dilution in deionized water, using ion-selective electrodes
(Nova Statprofile 5, Nova Biomedical, Waltham, MA). Urine
Cl2 concentration ([Cl2]) was measured by coulometric titration (Buchler-Cotlove chloridometer, Buchler Instruments,
Fort Lee, NJ). Plasma and urine creatinine concentrations
were determined with the use of an enzymatic fluorometric
technique (5) after urine was first diluted 49:1 (20 µl in 980 µl
H2O) in deionized water. Differences between duplicate measurements for the assays were 0.3 6 0.1 meq/l for [Na1], 0.6 6
0.2 meq/l for Cl2, 0.02 6 0.02 meq/l for [K1], 0.2 6 0.2 meq/l
for [Ca21] and 0.3 6 0.2 mmol/l for creatinine and phosphate
concentrations.
Urine TA minus bicarbonate concentration ([TA2HCO2
3 ])
was determined by using a double titration procedure (20) as
described previously (28). Briefly, immediately after collection, a 15-ml sample of urine was acidified to below pH 5 with
20 µl of concentrated (60%) nitric acid. The titration was
performed on 1.0-ml urine samples by using pH and reference
electrodes (MI-406, MI-403, Microelectrodes, Londonderry,
NH) with a digital pH meter (PHM 73, Radiometer, Copenhagen, Denmark). Humidified room air (23.6 6 0.1°C) was
bubbled through the urine samples for 15 min to complete the
removal of HCO2
3 from solution. A digital micrometer syringe
(model S4200A, Roger Gilmont Instruments, Great Neck,
NY) was used to dispense 0.1 N NaOH to titrate the sample
back to the corresponding arterial plasma pH.
Calculations. Urine [TA2HCO2
3 ] was calculated after Hills
(20)
542
RENAL RESPONSES TO KHCO3 AND NAHCO3 LOADING
Urine [SID] was calculated as the sum of the strong cations
minus the sum of the strong anions (35)
[SID] 5 ([Na1] 1 [K1] 1 [Ca21]) 2 ([Cl2] 1 [lactate2])
(2)
Urine [Atot] and [HCO2
3 ] were calculated according to the
following equation, which is consistent with the mass action
equilibria and electroneutrality of solutions (35)
[H1]4 1 (KA 1 [SID]) 3 [H1]3
1 5([SID] 2 [Atot])(KA) 2 (KC 3 PCO2 1 K8w)6 3 [H1]2
2 [(KC 3 PCO2 1 K8w)(KA) 1 (K3 3 Kc 3 PCO2)] 3 [H1]
(3)
2 KA 3 K3 3 KC 3 PCO2 5 0
Plasma acid-base state, ions, and aldosterone. In the
NaHCO3 trial, plasma [H1] decreased by 7.8 6 3 neq/l
by 60 min, compared with a 5.9 6 1.6 neq/l decrease in
the KHCO3 trial (Table 1). In both trials, plasma [H1]
remained significantly lower than initial levels until
the end of the experiment. Within 30 min of the
beginning of NaHCO3 ingestion, arterial plasma
[HCO2
3 ] increased and remained elevated until the end
of the experiment. In the KHCO3 trial, arterial plasma [
HCO2
3 ] increased above initial levels by 80 min and
returned toward initial values by 120 min. Detailed
responses and interpretation for both arterial and
venous plasma have been published (27).
In the NaHCO3 trial, arterial plasma [Na1] and [Cl2]
did not change (Table 1); in contrast, in the KHCO3 trial
both [Na1] and [Cl2] significantly decreased between
100 and 150 min, with [Na1] remaining depressed until
the end of the experiment. Plasma [K1] and aldosterone
did not change in the NaHCO3 trial (Fig. 1). In the
KHCO3 trial, plasma [K1] peaked at 7.17 6 0.13 meq/l
at 110 min and then slowly decreased to 5.1 6 0.8 by
270 min. The increase in plasma aldosterone concentration paralleled that of plasma [K1], with aldosterone
concentration exceeding 1 µmol/l between 90 and 150
min.
Water balance, GFR, and UFR. In the NaHCO3 trial,
there was no change in plasma volume during the
ingestion period, and then plasma volume progressively increased, peaking at 7.5 6 2.0% above initial
levels at 210 min (Table 1). In contrast, in the KHCO3
trial, ingestion of the solution resulted in an immediate
and pronounced decrease in plasma volume that reached a
Table 1. Plasma ions and percent change in plasma volume before (time 0), during (20, 40, and 60 min),
and after ingestion of NaHCO3 and KHCO3
Time, min
Preingestion
Trial
[Na1 ] NaHCO3
[Na1 ] KHCO3
[Cl2 ] NaHCO3
[Cl2 ] KHCO3
[HCO2
3 ] NaHCO3
[HCO2
3 ] KHCO3
[H1 ] NaHCO3
[H1 ] KHCO3
%dPV NaHCO3
%dPV KHCO3
0
143.1
6 0.6
143.2
6 0.2
106.3
6 0.3
106.1
6 0.6
26.0
6 0.4
24.2
6 1.1
37.6
6 0.6
39.3
6 0.9
0.0
0.0
Ingestion
20
40
Postingestion
60
80
100
120
150
180
144.8 146.0
146.8
146.1
146.5
146.5
145.1
145.5
6 0.7 6 0.8
6 0.8
6 0.6
6 0.8
6 0.9
6 0.5
6 0.4
142.8 142.4
142.3†
141.5†
140.9*† 140.9*† 141.3*† 141.1*†
6 0.3 6 0.3
6 0.2
6 0.3
6 0.6
6 0.4
6 0.3
6 0.2
105.9 104.5
104.3
102.9
103.7
102.4
104.1
103.5
6 0.4 6 0.8
6 0.9
6 0.9
6 0.9
6 1.3
6 0.9
6 0.8
106.3 105.7
105.3
104.7
105.0
104.4*
104.5*
104.9
6 0.5 6 0.6
6 0.7
6 0.7
6 0.5
6 0.4
6 0.3
6 0.3
27.9
29.8*
32.0*
32.5*
32.0*
31.7*
31.2*
30.3*
6 0.4 6 0.4
6 0.6
6 0.8
6 0.6
6 0.7
6 1.3
6 1.8
24.8†
26.6*†
27.3*†
28.4*†
29.9*†
27.3*†
27.2*†
25.8*†
6 0.9 6 0.8
6 0.9
6 0.9
6 1.1
6 0.6
6 0.5
6 0.8
36.0
33.1*
32.3*
32.4*
33.8*
34.3*
34.0*
36.0*
6 0.5 6 0.6
6 0.5
6 1.0
6 0.6
6 0.8
6 0.5
6 0.7
38.5† 38.0†
35.7*†
35.5*†
34.1*
34.7*
36.0*
36.4*
6 0.6 6 0.7
6 1.8
6 0.7
6 1.8
6 1.0
6 0.7
6 0.8
22.3
0.2
21.7
2.2
3.3
2.3
6.8*
6.6*
6 1.5 6 1.3
6 1.9
6 2.6
6 1.7
6 2.2
6 2.4
6 1.8
21.2 25.8†
210.2*† 211.2*† 213.4*† 214.9*† 210.5*† 210.8*†
6 0.7 6 1.5
6 2.7
6 3.3
6 2.5
6 1.7
6 4.9
6 3.5
210
240
270
145.3
6 0.5
140.9*
6 0.3
104.4
6 0.6
104.7
6 0.3
29.2*
6 0.6
25.9*†
6 1.4
35.0*
6 0.9
36.0*
6 0.6
7.5*
6 2.0
28.6*†
6 2.6
145.1
6 0.6
141.4*
6 0.3
103.4
6 1.3
105.1
6 0.6
30.1*
6 0.5
26.4*†
6 1.5
34.7*
6 0.6
35.4*
6 1.3
5.6*
6 1.8
25.8*†
6 3.9
144.3
6 0.8
141.5*
6 0.5
103.0
6 1.1
104.7
6 0.5
30.3*
6 0.5
25.3*†
6 1.4
35.6*
6 0.4
37.3*
6 0.5
5.9*
6 2.4
24.5*†
6 2.8
Values are means 6 SE; n 5 5. Ion concentrations are in meq/liter. %dPV, percent change in plasma volume. Brackets denote concentration.
* Significantly different from time 0, P # 0.05; † KHCO3 mean significantly different from NaHCO3 mean, P # 0.05.
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
where at 37°C the constants have the following values: KA 5
1.58 3 1027 eq/l (the dissociation constant for the phosphate
buffer system; Ref. 18), KC 5 2.45 3 10211 (eq/l)2/mmHg (35),
K3 5 6.0 3 10211 eq/l (11) and K8W 5 4.4 3 10214 (eq/l)2 (18).
Urine PCO2 was assumed to equal arterial plasma PCO2 for the
purposes of these calculations (see Ref. 28).
The validity of the physicochemical method was verified by
calculating urine [Atot] from measured urine pH, [SID], a
constant PCO2 of 40 Torr, and a KA for [Atot] of 1.58 3 1027 eq/l.
Linear regression analysis of [Atot] vs. [TA2HCO2
3 ] yielded
the significant (P , 0.0001; r2 5 0.806) relation
[Atot] 5 20.7950 6 0.0672 [TA2HCO2
3 ]. This calculation also
showed that calculated urine [HCO2
3 ] approximated [Atot] at
each time point in both trials. Consistent with the law of
electroneutrality, the sum of [Atot] and [HCO2
3 ] equaled [SID]
within 3.8 6 1.1%.
Statistics. All values are means 6 SE. Two-way ANOVA
with repeated measures was used to analyze data with
respect to time and treatment. When a significant F ratio was
obtained, the Student-Newman-Keuls method was used to
compare means. Statistical significance was accepted at
P # 0.05.
RESULTS
RENAL RESPONSES TO KHCO3 AND NAHCO3 LOADING
nadir of 214.9 6 1.7% below initial levels at 120 min;
this was followed by a slow, significant partial recovery. Water balance summaries (Table 2) are based on
estimates of complete intestinal absorption of the solutions by 270 min (see Ref. 27). The net increase in total
body water in the NaHCO3 trial was 555 6 119 ml,
compared with complete restoration of fluid balance
(25 6 83 ml) in the KHCO3 trial. In the NaHCO3 trial,
the ECF compartment was in positive fluid balance,
consistent with the retention of Na1 in the ECF compartment (27). In contrast, in the KHCO3 trial, ECFV
was estimated to be in negative balance by about 800
ml, consistent with a net movement of water into cells
(27).
In the NaHCO3 trial, initial GFR was 90 6 18 ml/min
and did not change throughout the experiment (Fig.
2A). In contrast, in the KHCO3 trial, GFR increased
rapidly from 71 6 16 ml/min (time 0) to 306 6 45
ml/min by 60 min and remained elevated until 120 min.
In the NaHCO3 trial, UFR increased two- to threefold
from 0.6 6 0.2 ml/min (time 0) to 2.6 6 0.4 ml/min
between 80 and 120 min (Fig. 2B). After KHCO3
ingestion, UFR increased up to eightfold greater than
initial within 80 min and remained significantly greater
than in the NaHCO3 trial until 150 min.
Sodium. In the NaHCO3 trial, 270 min after ingestion of 280 meq of Na1 was begun, 10% of ingested Na1
remained in the plasma compartment, 46% remained
in the interstitial fluid compartment, and renal UNa1V
accounted for 30% of ingested Na1 (Table 3).
In the NaHCO3 trial, urine [Na1] increased twofold
between the end of ingestion (123 6 27 meq/l at 60
min) and 270 min (255 6 14 meq/l). UNa1V was threeto fourfold greater than initial values between
80 and 180 min, and it remained elevated (298 6 14
µeq/min) at 270 min compared with initial levels
Table 2. Water balance with NaHCO3 and KHCO3
ingestion at 120 and 270 min
NaHCO3
Fluid ingested,
ml
Saline infused,
ml
Blood sampling,
ml
Urine volume,
ml
DPV, ml
DISFV, ml
DECFV, ml
DTBW, ml
KHCO3
120 min
270 min
936 6 70
936 6 70
915 6 59
915 6 59
251 6 36
434 6 79
260 6 29
438 6 66
240
400
207 6 20
69 6 57
311 6 255
380 6 311
740 6 93
120 min
240
270 min
400
415 6 36
509 6 45*
928 6 50*
173 6 67
2453 6 79* 2144 6 86*
780 6 303 22,039 6 357* 2650 6 389*
954 6 370 22,493 6 436* 2794 6 475
555 6 119
426 6 78*
25 6 83*
Values are means 6 SE; n 5 5. Fluid ingested, fluid volume
accompanying the NaHCO3 or KHCO3 dose; saline infused, measured
from iv drip at 270 min; blood sampling, estimated blood volume loss
due to blood sampling; urine volume, cumulative urine volume
produced. DPV, change in arterial plasma volume; DISFV, change in
interstitial fluid volume; DECFV, change in extracellular fluid volume; DTBW, change in total body water. The change in ECFV was
partitioned between the changes in PV and ISFV. Minus sign
indicates decrease in volume. * KHCO3 mean significantly different
from NaHCO3 mean, P , 0.05.
Fig. 2. Glomerular filtration rate (GFR, A) and urine flow rate (B)
before (up to 0 min), during (1–60 min), and after (61–270 min)
ingestion of NaHCO3 (j) or KHCO3 (r) at a dose of 3.57 mmol/kg
body mass. Urine flow rate in the KHCO3 trial was significantly
(P # 0.05) greater than in the NaHCO3 trial between 60 and 180 min.
GFR in the KHCO3 trial was significantly (P # 0.05) greater than in
the NaHCO3 trial between 60 and 120 min. Hatched bars indicate the
60-min period of HCO3 ingestion. Values are means 6 SE; n 5 5.
* Significantly different (P # 0.05) from preingestion (220 and 0 min).
V Significant difference between treatments.
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
Fig. 1. Plasma K1 concentration ([K1], solid symbols) and aldosterone concentration ([aldosterone], open symbols) before (up to 0 min),
during (1–60 min), and after (61–270 min) ingestion of NaHCO3
(squares) or KHCO3 (circles) at a dose of 3.57 mmol/kg body mass.
Hatched bar indicates 60-min period of HCO3 ingestion. Values are
means 6 SE; n 5 5. * [K1] and aldosterone concentration significantly
increased (P # 0.05) compared with preingestion (220 and 0 min).
V [K1] and [aldosterone] significantly greater in KHCO trial com3
pared with NaHCO3 trial.
543
544
RENAL RESPONSES TO KHCO3 AND NAHCO3 LOADING
Table 3. Na1 balance with NaHCO3 ingestion and K1
balance with KHCO3 ingestion at 270 min
Amount ingested
DPlasma content
DISF content
DECF content
Tissue flux
Renal excretion
Other effectors
Total
Na1 Balance With
NaHCO3 Ingestion
K1 Balance With
KHCO3 Ingestion
Na1,
meq
K1,
meq
280 6 21
28 6 9
128 6 42
157 6 51
30 6 158
84 6 7
28 6 157
%
Ingested
56 6 19
21 6 49
30 6 2
14 6 50
100 6 0
281 6 23
261
763
963
96 6 7
93 6 16
282 6 31
%
Ingested
361
37 6 5
34 6 6
27 6 9
100 6 0
(78 6 21 µeq/min; Fig. 3A). The fractional excretion of
Na1 increased fourfold by 120 min, and the mean
cumulative fractional excretion was approximately twofold greater than baseline after NaHCO3 ingestion
(Fig. 3B).
In the KHCO3 trial, urine [Na1] did not change and
was lower than in the NaHCO3 trial between 60 and
270 min (not shown). With KHCO3 ingestion, UNa1V
Fig. 3. Renal ion excretions (UNa1V, A; UK1V, C; and UCl2V, E) and fractional excretions of Na1 (B), K1 (D), and Cl2
(F) before (0 min), during (1–60 min), and after (61–270 min) ingestion of NaHCO3 (j) or KHCO3 (r) at a dose of
3.57 mmol/kg body mass. Hatched bars indicate the 60-min period of HCO3 ingestion. Values are means 6 SE; n 5 5.
* Significantly different (P #0.05) from preingestion (220 and 0 min). V Significant difference between treatments.
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
Values are means 6 SE; n 5 5. Not shown is ,65 meq of Na1
infused with the saline drip. DPlasma content, change in arterial
plasma content; DISF content, change in interstitial fluid content;
DECF content, change in extracellular fluid content. For changes in
Plasma, ISF, and ECF contents, minus sign indicates decrease.
Tissue flux, total whole body skeletal muscle influx (1) or efflux (2) of
Na1 or K1 (27); renal excretion, cumulative Na1 or K1 excretion;
other effectors (gastrointestinal tract, erythrocytes), changes in ECF
content not accounted for. % Ingested, percent ingested Na1 or K1
accounted for by tissue flux, renal excretion, and other effectors.
increased rapidly and was greater than initial levels
between 60 and 120 min and then declined toward
initial values by 180 min (Fig. 3A). Between 60 and 120
min, UNa1V was greater in the KHCO3 than in the
NaHCO3 trial. There was no significant change in the
fractional excretion of Na1 (Fig. 3B).
Potassium. In the NaHCO3 trial, urine [K1] remained unchanged at ,104 6 16 meq/l (not shown);
excretion and the fractional excretion of K1 also did not
change (Fig. 3, C and D). In contrast, in the KHCO3
trial, urine [K1] increased rapidly from 77 6 19 meq/l at
120 min to 241 6 37 meq/l at 270 min. This increase
was accompanied by a sixfold increase in UK1V between
120 and 180 min (Fig. 3C) and an elevated K1 fractional
excretion to 71 6 12% of the filtered load in the last 30
min of the experiment (Fig. 3D). In the KHCO3 trial, of
the 281 meq of K1 ingested, the increase in ECF K1
content only accounted for 3%, whereas net K1 flux into
tissues accounted for 37% (Table 3). Over the 270 min of
the experiment, increased (above basal) UK1V accounted for 26 6 5% of the ingested K1 or 76% of the
total cumulative UK1V of 93 6 16 meq (Table 3).
Chloride. In the NaHCO3 trial, urine [Cl2] decreased
from 186 6 16 meq/l at 0 min to 40 6 7 meq/l at 120
min, with no change in Cl2 excretion or fractional
excretion (Fig. 3, E and F). In contrast, in the KHCO3
trial, urine [Cl2] decreased by only ,80 meq/l, from
197 6 29 meq/l at 0 min to 119 6 12 meq/l at 80 min,
returning to 173 6 15 meq/l at 270 min. Despite the
decrease in urine [Cl2], the marked increases in UFR
and GFR resulted in a significantly increased renal Cl2
RENAL RESPONSES TO KHCO3 AND NAHCO3 LOADING
Fig. 4. Urine strong ion difference concentration ([SID]) (A), urine
pH (B), and urine [H1] (C) before (0 min), during (1–60 min), and
after (61–270 min) ingestion of NaHCO3 (j) or KHCO3 (r) at a dose
of 3.57 mmol/kg body mass. Dashed line in C represents relationship
between [H1] and [SID] in the absence of CO2 in the solution (37).
Urine [SID] (A) in NaHCO3 trial was significantly (P #0.05) greater
than in KHCO3 trial from 80 to 270 min. Monoexponential relationship between urine [SID] and urine [H1] is described by the equation
[H1] 5 59.168 6 4.223exp(2[SID] 3 0.0122 6 0.00154) (r2 5 0.670; P , 0.001;
n 5 54). Hatched bars indicate the 60-min period of HCO3 ingestion.
Values are means 6 SE; n 5 5. * Significantly different (P # 0.05)
from preingestion (220 and 0 min). V Significant difference between
treatments.
and rapid decrease in UTA2HCO23 V (Fig. 5B) represented
a pronounced net excretion of titratable base between
80 and 180 min; there was no difference between trials.
In both trials, urine [NH1
4 ] decreased sevenfold by
100 min, with no difference between trials (Fig. 5D)
UNH14 V did not change in either trial (Fig. 5D). Urine
net acid concentration (Fig. 5E) and net acid excretion
(Fig. 5F) were quantitatively similar to that for TA2
V formed only a small (1–3%)
HCO2
3 , because UNH1
4
proportion of total acid excretion. In the NaHCO3
trial, an amount equivalent to 23 6 3% of the ingested
base was excreted, whereas in the KHCO3 trial
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
excretion between 60 and 120 min (Fig. 3E). Cl2
fractional excretion did not change significantly in
either trial, although Cl2 fractional excretion was
greater in the KHCO3 trial than in the NaHCO3 trial
between 80 and 270 min (Fig. 3F).
Calcium, Pi, and lactate. In both trials, urine [Ca21]
decreased two- to threefold from 3.6 6 0.6 meq/l at 0
min to between 1 and 2 meq/l from 80 and 270 min, with
no difference between trials. Renal Ca21 excretion and
Ca21 fractional excretion did not change from initial
values (3 6 1 µeq/min and 1.7 6 0.7%, respectively) in
either trial, with no difference between treatments.
In the NaHCO3 trial, urine [Pi] remained unchanged
at about 24 6 4 mmol/l. In the KHCO3 trial, urine [Pi]
decreased threefold (from 26 6 13 mmol/l) at time 0 to
4.3 6 1.6 mmol/l within 80 min and remained low
thereafter. Pi excretion did not change (20 6 9 µmol/
min) in either treatment, with no difference between
trials.
Urine lactate2 concentration was initially 0.02 6
0.01 meq/l in both trials, and in both trials it approached zero by 270 min (not shown).
[SID], pH, and [H1]. In the NaHCO3 trial, urine
[SID] increased rapidly until 80 min and then increased slowly until the end of the experiment (Fig. 4A).
In the KHCO3 trial, urine [SID] increased gradually
and was greater than initially between 180 and 270
min, but it remained 125–175 meq/l lower than in the
NaHCO3 trial.
In both trials, the magnitude and time course of
increase in urine pH were similar, with nadirs reached
at 80 min; there were no differences between trials (Fig.
4B). In the NaHCO3 trial, urine [H1] decreased significantly from 354 6 283 neq/l (time 0) to 6.8 6 0.4 neq/l at
270 min, compared with 1,414 6 1,319 neq/l (time 0)
and 13.7 6 3.9 neq/l at 270 min in the KHCO3 trial.
Urine [SID] (meq/l) was a good predictor of urine [H1]
(neq/l), as shown by the monoexponential curve fit to
the data (Fig. 4C). The lower dotted line in Fig. 4C
represents the relationship between [H1] and [SID] in
the absence of CO2 and weak acids in the solution (35).
The difference between the dotted line and the experimental data represents the acidification contributed by
urine PCO2. The dashed line represents the relationship
between [H1] and [SID] when solution PCO2 is 40 Torr
with no weak acids present ([Atot] 5 0): [H1] 5 KC 3
PCO2/[SID], with units for [SID] in eq/l (35). The dashed
line fits the data well, supporting the assumption that
urine PCO2 was similar to arterial PCO2. The relationship also shows that acidification by CO2 was pronounced at low and negative urine [SID] (before bicarbonate ingestion) and minor when [SID] was .100
meq/l (after the start of bicarbonate ingestion). The
small amount of weak acids present in urine did not
contribute substantially to the relationship (35).
TA, NH41, and net acid excretion. In both trials, there
were large and rapid decreases in urine [TA2HCO2
3 ],
UTA2HCO23 V, and net acid excretion (Fig. 5). Urine [TA2
HCO2
3 ] was significantly decreased from initial values
by 80 min in both trials, with similar magnitudes and
time course of change (Fig. 5A). The associated large
545
546
RENAL RESPONSES TO KHCO3 AND NAHCO3 LOADING
the base excreted accounted for 25 6 3% of that
ingested.
DISCUSSION
This appears to be the first study to compare the
acute renal responses to large, equimolar doses (3.57
mmol/kg body mass) of ingested NaHCO3 and KHCO3
in humans and in mammals in general. The dose of
NaHCO3 used is sufficient to be of ergogenic benefit
(19), but this dose of KHCO3 was three to four times
greater than that used in previous studies. Earlier
studies focused on longer term responses and therefore
missed the rapidity with which the kidneys respond to
large fluid and electrolyte loads. Furthermore, the
ingested Na1 and K1 resulted in marked differences in
renal fluid and ion excretion, consistent with differences in intestinal Na1 and K1 absorption and differences in the distribution of these cations within body
fluid compartments. It is noteworthy that, despite such
differences, there was little difference in the renal
excretion of base between the two treatments.
Urine acid-base status. Stewart (35) has mathematically described the physicochemical relationships among
independent variables contributing to [H1] and [HCO2
3]
in physiological solutions. The independent determinants of [H1] and [HCO2
3 ] in urine are the concentrations of strong (fully dissociated) ions represented by
the [SID], the total concentration of weak acids and
base represented by [Atot], and PCO2 (28, 35). In healthy,
resting humans, the main determinants of [SID] are
[Na1], [K1], [Cl2], and [Ca21]. The ions NH1
4 (pK 9.26),
play
minor
roles
Mg21, lactate2 (pK 5 3.8), and SO22
4
and largely cancel each other with respect to charge. In
urine, the sum of the charges of strong ions, the [SID],
thus provides an index of total acid excretion (8). The
main determinants of [Atot] are weak organic anions
[the most abundant of which are citrate and some
amino acids (8)] and Pi. [Atot] is equivalent to the TA
minus bicarbonate term of classical renal physiology.
TA refers to the amount of secreted acid titrated by
nonvolatile buffers, i.e., weak acids, in the tubular
lumen and subsequently excreted as fixed acid. Pi with
an apparent pK8 5 6.8 accounts for most of the TA
content of the urine (20, 35). In classical renal physiology, acid is excreted in the urine as NH1
4 and TA, so that
2
net acid excretion equals NH1
4 1 TA 2 HCO3 excretion.
An unexpected result was the marked similarity in
renal base excretion in both trials despite the marked
differences in rate and magnitude of plasma fluid and
ion disturbances and differences in renal water and ion
handling (see below). The rate of renal base excretion,
however, paralleled that of arterial alkalinization, of
which changes in plasma [SID] were the primary
determinant (27). The primary determinant of changes
in urine [H1] in both trials was the change in urine
[SID]. In general, KHCO3 ingestion resulted in lower
urine [Na1], [K1], and [Ca21] than in the NaHCO3 trial.
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
2
Fig. 5. Urine titratable acid minus HCO3 concentration ([TA 2 HCO2
3 ], A), renal TA 2 HCO3 excretion (UTA2HCO3V;
1
1
1
B), [NH4 ] (C), NH4 excretion (UNH4 V; D), net acid concentration ([net acid], E), and net acid excretion (UNetAcidV; F)
before (0 min), during (1–60 min), and after (61–270 min) ingestion of NaHCO3 (j) or KHCO3 (r) at a dose of 3.57
mmol/kg body mass. Negative values represent net base excretion. Hatched bars indicate 60-min period of HCO3
ingestion. Values are means 6 SE; n 5 5. * Significantly different (P # 0.05) from preingestion (220 and 0 min).
V Significant difference between treatments.
RENAL RESPONSES TO KHCO3 AND NAHCO3 LOADING
fourfold increases in GFR and excretion of Na1, Cl2,
and K1.
The magnitude of increase in GFR in response to
KHCO3 ingestion appears to be without precedence in
the literature. One study on humans reported no
change in GFR in the second hour after oral ingestion of
a small amount (1 mmol/kg body mass) of KCl (37).
Modest increases (119 ml/min) in GFR in sheep given
KCl have been reported (4), whereas in rats a decrease
in GFR occurring one or more hours after K1 loading
has been reported (6, 7, 40). Also, earlier studies did not
report GFR until at least 1 h after K1 loading. From
Fig. 2, it is evident that the increase in GFR was rapid
and transient and had returned to preingestion values
within 210 min after KHCO3 ingestion was completed.
It is difficult to provide a mechanistic explanation for
the fourfold increase in GFR in the absence of measures
of blood pressure, peripheral vascular resistance, and
heart rate. It is not likely that KHCO3 ingestion was
associated with marked increases in blood pressure,
given that PV decreased markedly, nor with increases
in vascular resistance, given that elevated plasma [K1]
has vasodilatory effects. The rapidity and magnitude at
which the hyperkalemia ensued, compared with earlier
studies (9, 30, 37), may have induced indirect effects by
increasing renal blood flow and thereby contributing to
an increase in net filtration pressure and perfusion
pressure at the level of the juxtaglomerular apparatus.
However, it is unknown if the changes in tubular water
and ion handling were due to the increase in GFR or to
the increase in filtered K1 (7), and further study is
required to understand the mechanistic relationships
responsible for these observations.
Na1 excretion. In the NaHCO3 trial, the rapid and
large increases in urine [Na1], UNa1V, and fractional
excretion accounted for 24% of ingested Na1, with the
bulk of the Na1 remaining in the ECF compartment
(27). The increase in UNa1V can be attributed to the
modest increases in GFR and increased Na1 delivery
with decreased proximal tubule Na1 reabsorption. The
primary mechanism for reduced proximal tubule Na1
reabsorption is through inhibition of proximal tubular
Na1-K1-ATPase (2, 22), resulting in increased Na1
(and water) delivery to the distal tubules. There is also
evidence that acute, transient elevations in dietary
Na1 induce a natriuresis by a dopamine-mediated
decrease in proximal tubule Na1-K1-ATPase activity (1).
In the KHCO3 trial the hyperkalemia was associated
with increased renal UNa1V despite the 15% reduction
in plasma volume and decrease in plasma [Na1]. Hyperkalemia has been shown to inhibit Na1 and water
absorption by the proximal tubule, resulting in a
diuresis and natriuresis (6, 7) without change in plasma
renin (30, 37, 38) and atrial natriuretic peptide (37).
Increased GFR was associated with elevated distal
tubule Na1 delivery with minimal increase in Na1
fractional excretion. There appears to be minimal evidence for decreased Na1 reabsorption in the diuretic
and natriuretic responses to KHCO3 ingestion. This
finding is in contrast to the natriuresis that occurs with
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
In addition, urine [Cl2] was greater throughout the
KHCO3 trial. Urine [SID] increased by ,175 meq/l
greater in the NaHCO3 trial than in the KHCO3 trial,
and this balanced the decrease in urine [TA2HCO3]
and net acid concentration. In general, KHCO3 ingestion resulted in lower urine [Na1], [K1], and [Ca21], and
higher [Cl2] than in the NaHCO3 trial. Changes in CO2
did not contribute substantially to urine [H1] (see Fig.
4C), consistent with the fact that arterial, and hence
renal, PCO2 (not shown) changed little in both trials.
An increase in urine [Atot] also contributed to the
measured urine [H1]. In both trials large decreases in
[TA2HCO2
3 ] were largely independent of changes in
urine [Pi], with urine [Pi] only accounting for a small
fraction of [TA2HCO2
3 ]. As previously described (8),
this suggests that there was an increased excretion of
weak acids and bases, but their identity and contributions are not known in the present study. The contribution of NH1
4 to net acid excretion was small (1–3%),
consistent with a previous report of decreased UNH14 V
after KHCO3 and KCl ingestion (37).
In the present study, and in that by Van Buren et al.
(37), much of the decrease in net acid excretion would
be classically described as due to increased excretion of
HCO2
3 . This characterization is consistent with an
apparent 40% reduction in tubular HCO2
3 reabsorption
in rats with acute metabolic alkalosis (33) and with
studies showing that loading with KHCO3, NaHCO3,
KCl, and NaCl results in decreased renal HCO2
3 reabsorption and decreased urine [H1] in dogs (29) and
humans (2, 37). In addition, there is also evidence that
electrogenic proton secretion is reduced in rat nephrons
as soon as 3 h after the onset of NaHCO3 loading (24).
This is consistent with the large, transient decreases in
TA (37) and TA2HCO2
3 (present study) excretion observed 1–3 h after K1 loading in humans.
Water balance. Ingestion of NaHCO3 resulted in a
prolonged retention of fluid with total cumulative urine
volume accounting for only 44% of ingested volume. An
estimated 56% of ingested Na1 remained in the ECF
compartment at the end of the experiment (Table 3)
consistent with the nearly 1-liter expansion of ECFV
(27). Increases in the delivery of Na1, K1, and base, i.e.,
increased osmolal clearance, to the distal tubules,
without concomitant change in GFR, suggest that
decreased tubular fluid reabsorption occurred in association with increased distal tubule Na1 delivery. The
increase in UFR, however, was largely normalized 2 h
after ingestion of the solution, indicating that there
was a decreased sensitivity of the volume regulatory
system or some feed-forward mechanism to prevent
overadjustment in terms of water excretion and UNa1V.
In contrast, ingestion of KHCO3 resulted in a rapid
and pronounced decrease in plasma volume that was
attributed to an initial rapid net movement of fluid into
the proximal small intestine to bring intestinal contents toward plasma osmolarity, with subsequent absorption of water and K1 in more distal portions of the
small intestine (27). Despite the ,0.5-liter reduction in
plasma volume, UFR increased two- to threefold more
than in the NaHCO3 trial, and there were rapid,
547
548
RENAL RESPONSES TO KHCO3 AND NAHCO3 LOADING
tubule and cortical collecting duct also stimulates K1
secretion by these segments (16, 26).
Cl2 excretion. In the NaHCO3 trial, as may be
expected with increased HCO2
3 delivery to the tubules,
there occurred a 146 meq/l decrease in urine [Cl2] and a
modest increase (not statistically significant but appearing to be of physiological importance) in tubular Cl2
reabsorption (Fig. 3F). This response is similar to that
seen occurring after acute lactate2 loading caused by
high-intensity exercise (28); there appears to be a
preferential reabsorption of Cl2 over HCO2
3 and lactate2. Overall, though, the NaHCO3 load had only a
small effect on renal Cl2 transport, similar to the
absence of effect of NaCl loading on renal Cl2 excretion
(17, 37).
The renal Cl2 response to KHCO3 ingestion, however,
is in marked contrast to that seen in the NaHCO3 trial.
Cl2 excretion markedly increased equimolar with the
increased excretion of Na1, resulting in similar cumulative excretions of Cl2 and Na1. Similar responses to
KHCO3 loading in humans have previously been reported (37), and increased HCO2
3 delivery has been
suggested to reduce net Cl2 reabsorption by the proximal tubules and increase Cl2 excretion (37). The absence of a similar result in the NaHCO3 trial may be
explained by the absence of the rapid and very large
increase in GFR seen in the KHCO3 trial. Thus the
marked increases in both Cl2 and Na1 excretion seen
after KHCO3 ingestion appear to be associated with
mechanisms for acute regulation of plasma [K1] (10, 25,
37). The decrease in Cl2 excretion late in the experiment (180–270 min) was similar to that seen for Na1
and is consistent with Na1-dependent Cl2 absorption
mechanisms responding to elevated plasma aldosterone concentration (10, 17).
Renal Ca21 and Pi regulation. The ingestion of
NaHCO3 and KHCO3 did not significantly affect renal
Ca21 and Pi excretion. In both trials, the decreases in
urine [Ca21] and [Pi] were due solely to the accompanying diuresis, consistent with the absence of significant
change in plasma Ca21 content (27).
Implications for exercise performance. NaHCO3 loading is widely practiced in human and equine athletic
competition as a means of reducing the severity of
extra- and intracellular acidosis resulting from the
performance of high-intensity exercise (19). Indeed, the
dose of NaHCO3 administered in the present study
consistently results in performance-enhancing effects
in humans (19). It is also noteworthy that NaHCO3
loading resulted in an ,1-liter expansion of plasma
volume that persisted in excess of 3 h after ingestion of
the solution. This result indicates that the gastrointestinal tract may be used as a reservoir of readily
available water and Na1 for maintaining extracellular
volume during prolonged exercise. On the negative
side, there was a twofold increase in UFR during the
first 2 h after ingestion that may compromise exercise
performance.
High-intensity exercise results in the rapid loss of K1
from contracting skeletal muscle, and the resulting
decrease in intracellular [K1] is thought to contribute
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
KCl ingestion (38), indicating that the accompanying
anion plays a role in modulating tubular excretion and
reabsorption of Na1 (22). The later decline in renal
UNa1V (from 120 min onward) is attributed to the
strong antinatriuretic action of aldosterone and the
ensuing increase in tubular Na1 reabsorption (31, 37,
38, 41). This effect and associated decreases in GFR,
UFR, and Cl2 excretion are consistent with the effects
of aldosterone occurring 1–2 h after it begins to increase in the blood (12, 41). It is concluded that a
K1-stimulated natriuresis was responsible for the observed hyponatremia.
K1 excretion. NaHCO3 ingestion resulted in a small,
although not statistically significant, increase in UK1V,
a tendency (Fig. 4) that was associated with a modest
hypokalemia. A tendency (P , 0.1) toward increased
UK1V is consistent with increased delivery and net
reabsorption of Na1, but not Cl2, at the distal tubules;
this tendency establishes a negative electrochemical
gradient that favors modest increases in UK1V (16, 39).
Alkalosis has also been reported to stimulate Na1-K1ATPase-mediated uptake of K1 into principal cells,
resulting in enhanced K1 secretion (34). An increased
delivery of HCO2
3 to the distal nephron, in the presence
of increased aldosterone, may also stimulate UK1V
during alkalosis (34).
In the KHCO3 trial, increased excretion of K1 above
basal levels accounted for 26 6 5% of ingested K1. Over
the course of the experiment, total UK1V accounted for
34 6 6% (about 1.2 mmol/kg) of ingested K1, representing about 17% of the normal kidneys’ daily capacity of 7
mmol/kg (31). The incomplete renal excretion of all
ingested K1 agrees with the observation that about
64% of ingested K1 had entered cells by 270 min and
was thus removed from the ECF and the circulatory
system (27).
The rapid, fivefold increase in renal UK1V was biphasic in time course, peaking at 90 min before decreasing
(Fig. 3C), and paralleled increases in plasma [K1] and
aldosterone concentration (Fig. 1). The early rise in
UK1V preceded the kaliuretic action of increased plasma
aldosterone concentration but coincided with the increase in filtered K1 load, perhaps indicating a direct
kaliuretic effect of increased plasma [K1] (30). In
contrast, the fractional excretion of K1 progressively
increased over time, peaking at 71 6 12% of the filtered
load at 270 min, consistent with the peak effects of
aldosterone occurring 1–2 h after it increases (12, 41).
These results are consistent with previous studies of
K1 loading in humans (9, 30, 37, 38) and other animals
(3, 6, 7, 21, 32, 41). It is likely that elevated plasma [K1]
and aldosterone concentration independently contributed to the kaliuresis (9, 31) by increasing principal cell
Na1-K1-ATPase activity (34) and intracellular [K1] in
the distal tubule and cortical collecting duct (13, 15).
Also, the presence of anions that are relatively impermeant to distal tubule reabsorption (such as bicarbonate and sulphate) increases luminal electronegativity
and may have contributed to the increase in K1 secretion (39). An increased flow of fluid through the distal
RENAL RESPONSES TO KHCO3 AND NAHCO3 LOADING
This study was supported by the Natural Sciences and Engineering Research Council of Canada and the Medical Research Council of
Canada. G. J. F. Heigenhauser is a Career Investigator with the
Heart and Stroke Foundation of Ontario. L. C. Lands is a Chercheurclinicien with the Fonds de la Recherché en Santé du Quebec.
Address for reprint requests and other correspondence: M. I.
Lindinger, Dept. of Human Biology & Nutritional Sciences, Univ. of
Guelph, Guelph, ON, Canada N1G 2W1 (E-mail: mlindinger.ns@
aps.uoguelph.ca).
Received 16 February 1999; accepted in final form 15 October 1999.
REFERENCES
1. Aperia, A., J. Fryckstedt, U. Holtback, R. Belusa, X.-J.
Cheng, A.-C. Eklof, D. Li, Z.-M. Wang, and Y. Ohtomo.
Cellular mechanisms for bi-directional regulation of tubular
sodium reabsorption. Kidney Int. 49: 1743–1747, 1996.
2. Batle, D. C., and N. A. Kurtzman. Renal regulation of acidbase homeostasis: integrated response. In: The Kidney: Physiology and Pathophysiology, edited by D. W. Seldin and G. Giebisch.
New York: Raven, 1985, p. 1539–1566.
3. Baylis, C., and W. J. O’Connor. The effect of plasma potassium
in determining normal rates of excretion of potassium in dogs.
Q. J. Exp. Physiol. 61: 145–157, 1976.
4. Beal, A. M., O. E. Budz-Olzen, R. C. Clark, R. B. Cross, and
T. J. French. Renal and salivary responses to infusion of
potassium chloride, bicarbonate, and phosphate in Merino sheep.
Q. J. Exp. Physiol. 58: 251–265, 1973.
5. Bergmeyer, H. U. (Editor). Methods of Enzymatic Analysis.
New York: Academic, 1974.
6. Braam, B., P. Boer, and H. A. Koomans. Tubuloglomerular
feedback and tubular reabsorption during acute potassium loading in rats. Am. J. Physiol. Renal Fluid Electrolyte Physiol. 267:
F223–F230, 1994.
7. Brandis, M., J. Keyes, and E. E. Windhager. Potassiuminduced inhibition of proximal tubular fluid reabsorption in rats.
Am. J. Physiol. 222: 421–427, 1972.
8. Brown, J. C., R. K. Packer, and M. A. Knepper. Role of
organic anions in renal response to dietary acid and base loads.
Am. J. Physiol. Renal Fluid Electrolyte Physiol. 257: F170–F176,
1989.
9. Calo, L., A. Borsatti, S. Favaro, and L. Rabinowitz. Kaliuresis in normal subjects following oral potassium citrate intake
without increased plasma potassium concentration. Nephron 69:
253–258, 1995.
10. Chen, P. Y., and A. S. Verkman. Sodium-dependent chloride
transport in basolateral membrane vesicles isolated from rabbit
proximal tubule. Biochemistry 27: 655–660, 1988.
11. Edsall, J. T., and J. Wymann. Biophysical Chemistry. New
York: Academic, 1958, vol. 1, p. 578–587.
12. Fanestil, D. D., and C. S. Park. Steroid hormones and the
kidney. Annu. Rev. Physiol. 43: 637–649, 1981.
13. Field, M. J., B. A. Stanton, and G. H. Giebisch. Differential
acute effects of aldosterone, dexamethasone and hyperkalemia
on distal tubular potassium secretion in the rat kidney. J. Clin.
Invest. 74: 1792–1802, 1984.
14. Gennari, F. J., and J. J. Cohen. Role of the kidney in
potassium homeostasis: lessons from acid-base disturbances.
Kidney Int. 8: 1–5, 1975.
15. Giebisch, G., and W. Wang. Potassium transport: from clearance to channels and pump. Kidney Int. 49: 1624–1631, 1996.
16. Good, D. W., and F. S. Wright. Luminal influences on potassium secretion: sodium concentration and fluid flow rate. Am. J.
Physiol. Renal Fluid Electrolyte Physiol. 236: F192–F205, 1979.
17. Greger, R. Chloride reabsorption in the rabbit cortical thick
ascending limb of the loop of Henle: a sodium dependent process.
Pflügers Arch. 390: 38–43, 1981.
18. Harned, H. S., and B. O. Owen. The Physical Chemistry of
Electrolyte Solutions (3rd ed.). New York: Van NostrandReinhold, 1958.
19. Heigenhauser, G. J. F., and N. L. Jones. Bicarbonate loading.
In: Perspectives in Exercise Science and Sports Medicine: Ergogenics-Enhancement of Performance Exercise and Sport, edited by
D. R. Lamb and M. H. Williams. Dubuque, IA: William C. Brown,
1991, vol. 4, p. 183–221.
20. Hills, A. G. Acid-Base Balance: Chemistry, Physiology, Pathophysiology. Baltimore, MD: Williams & Wilkins, 1973, p. 115–
126.
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
to skeletal muscle fatigue (see Ref. 27). The rationale
for providing subjects with oral KHCO3, at a dose
equivalent to the performance-enhancing effect of
NaHCO3, was twofold. The first was to provide a
similar magnitude alkalosis to offset exercise-induced
acidosis, with the idea that K1 being predominantly
intracellular would provide further protection against
intracellular acidosis. The second was to provide additional K1 to skeletal muscles so as to better maintain
intracellular [K1] and delay the onset of fatigue in the
face of contraction.
However, the results of the present study indicate
that KHCO3 loading should not be considered for the
enhancement of exercise performance. Ingestion of this
quantity of K1 poses safety concerns, both at rest and
particularly if subjects are contemplating exercise.
Ingestion of more than 150 mmol (about 16 g) of KHCO3
results in a rapid increase in plasma [K1] that may
require prompt treatment for hyperkalemia (see Ref.
27). Two hours after ingestion of KHCO3, subjects
maintained an average plasma [K1] of about 6 meq/l,
which, with high-intensity exercise, could result in
life-threatening increases in plasma [K1]. Also, it is
likely that the rapid and pronounced decrease in plasma
volume that occurred with KHCO3 ingestion may impose additional stress on the cardiovascular and intestinal systems during exercise.
Conclusions. Renal responses to ingested NaHCO3
and KHCO3 solutions are more rapid and of greater
magnitude than previously appreciated. In both trials,
excreted Na1 and K1 accounted for 25–40% of the
ingested ion and, in the KHCO3 trial, cumulative urine
volume over 270 min equaled the ingested volume load.
In both trials, the GFR and UFR responses were
transient and largely normalized by 210 min postingestion. The neural and cellular mechanisms for the rapid
up- and downregulation of GFR and UFR in the
NaHCO3 trial are not well understood. The rapidity of
the downregulation suggests either that, after some
time, the regulatory systems become tolerant of the
remaining fluid imbalance or that there may be feedforward mechanisms for preventing overadjustment of
fluid and Na1 balance. In the KHCO3 compared with
the NaHCO3 trial, renal ion excretions occurred with
greater rapidity, were of greater magnitude, and resulted in an increased osmotic clearance. In both trials,
increases in base excretion were sustained until the
end of the experimental period. It is also noteworthy
that, although the plasma acid-base disturbances had
markedly different physicochemical origins (27), the
renal acid-base responses were similar with respect to
excretion of base. The rapidity and magnitude with
which the kidneys responded to the NaHCO3 and
KHCO3 loading demonstrate their capacity, and physiological importance, for restoring body fluid homeostasis in the face of large, acute disturbances in water, ion,
and acid-base balance.
549
550
RENAL RESPONSES TO KHCO3 AND NAHCO3 LOADING
32. Rabinowitz, L., R. L. Sarason, C. Tanasovich, V. E. Mendel,
and R. P. Brockman. Effects of glucagon, insulin, proprionate,
acetate, and HCO3 on K excretion in sheep. Am. J. Physiol. Renal
Fluid Electrolyte Physiol. 246: F197–F204, 1984.
33. Seki, G., S. Coppola, K. Yoshitomo, B.-C. Burckhardt, I.
Samarzija, S. Muller-Berger, and E. Fromter. On the mechanism of bicarbonate exit from renal proximal tubular cells.
Kidney Int. 49: 1671–1677, 1996.
34. Silva, P., J. P. Hayslett, and F. H. Epstein. The role of
Na,K-activated adenosine triphosphatase in potassium adaptation: stimulation of enzymatic activity by potassium loading. J.
Clin. Invest. 52: 2665–2671, 1973.
35. Stewart, P. A. How To Understand Acid-Base: A Quantitative
Acid-Base Primer for Biology and Medicine. New York: Elsevier/
North-Holland, 1981.
36. Van Acker, B. A. C., G. C. M. Koomen, M. G. Koopman, R. T.
Krediet, and L. Arisz. Discrepancy between circadian rhythms
of inulin and creatinine clearance. J. Lab. Clin. Med. 120:
400–410, 1992.
37. Van Buren, M., T. J. Rabelink, H. J. M. Van Rijn, and H. A.
Koomans. Effects of NaCl, KCl and KHCO3 loads on renal
electrolyte excretion in humans. Clin. Sci. 83: 567–574, 1992.
38. Van Buren, M., H. J. M. Van Rijn, and H. A. Koomans.
Short-term indomethacin administration does not impair excretion of acute potassium load in humans. Eur. J. Clin. Invest. 22:
821–826, 1992.
39. Velazquez, H., F. S. Wright, and D. W. Good. Luminal
influences on potassium secretion: chloride replacement with
sulphate. Am. J. Physiol. Renal Fluid Electrolyte Physiol. 242:
F46–F55, 1982.
40. Young, D. B. Relationship between plasma potassium concentration and renal potassium excretion. Am. J. Physiol. Renal Fluid
Electrolyte Physiol. 242: F599–F603, 1982.
41. Young, D. B., R. E. McCaa, Y. J. Pan, and A. C. Guyton.
Effectiveness of the aldosterone sodium and potassium feedback
control system. Am. J. Physiol. 231: 945–953, 1976.
Downloaded from http://jap.physiology.org/ by 10.220.32.246 on June 16, 2017
21. Horisberger, J. D., and J. Diezi. Effects of mineralocorticoids
on Na1 and K1 excretion in the adrenalectomized rat. Am. J.
Physiol. Renal Fluid Electrolyte Physiol. 245: F89–F99, 1983.
22. Isozaki, T., H. Kumagai, M. Ohura, and A. Hishida. Natriuretic response to acute sodium chloride or sodium bicarbonate
infusions in humans. Miner. Electrolyte Metab. 21: 383–390,
1995.
23. Keith, N. M., A. E. Osterberg, and H. E. King. The excretion of
potassium by the normal and diseased kidney. Trans. Assoc. Am.
Physicians. 55: 219–222, 1940.
24. Khadouri, C., S. Marsy, C. Barlet-Bas, L. Cheval, and A.
Doucet. Effect of metabolic acidosis and alkalosis on NEMsensitive ATPase in rat nephron segments. Am. J. Physiol. Renal
Fluid Electrolyte Physiol. 262: F583–F590, 1992.
25. Kirchner, K. A. Effect of acute potassium infusion on loop
segment chloride reabsorption in the rat. Am. J. Physiol. Renal
Fluid Electrolyte Physiol. 244: F599–F605, 1983.
26. Kunau, R. T., M. L. Webb, and S. Borman. Characteristics of
the relationship between the flow rate of tubular fluid and
potassium transport to the distal tubule of the rat. J. Clin. Invest.
54: 1488–1495, 1974.
27. Lindinger, M. I., L. C. Lands, P. K. Pedersen, D. G. Welsh,
and G. J. F. Heigenhauser. Role of skeletal muscle in plasma
ion and acid-base regulation after NaHCO3 and KHCO3 loading
in humans. Am. J. Physiol. Regulatory Integrative Comp. Physiol.
276: R32–R43, 1999.
28. McKelvie, R. S., M. I. Lindinger, G. J. F. Heigenhauser, J. R.
Sutton, and N. L. Jones. Renal responses to exercise-induced
lactic acidosis. Am. J. Physiol. Regulatory Integrative Comp.
Physiol. 257: R102–R108, 1989.
29. Pitts, R. F., J. L. Ayer, and W. A. Schiess. Bicarbonate
reabsorption and changes in glomerular filtration. J. Clin.
Invest. 28: 35–43, 1949.
30. Rabelink, T. J., H. A. Koomans, R. J. Hene, and E. J.
Dorhout Mees. Early and late adjustment to potassium loading
in humans. Kidney Int. 38: 942–947, 1990.
31. Rabinowitz, L. Aldosterone and potassium homeostasis. Kidney Int. 49: 1738–1742, 1996.